Submit manuscript...
eISSN: 2378-315X

Biometrics & Biostatistics International Journal

Research Article Volume 7 Issue 5

Bootstrap confidence intervals for dissolution similarity factor f2

Mohammad M Islam,1 Munni Begum2

1Department of Mathematics, Utah Valley University, USA
2Department of Mathematical Sciences, Ball State University, USA

Correspondence: Mohammad M Islam, Department of Mathematics, Utah Valley University, 800 W University Pkwy, Orem, UT 84058, USA, Tel 1801 8636 430

Received: August 16, 2018 | Published: September 18, 2018

Citation: Islam MM, Begum M. Bootstrap confidence intervals for dissolution similarity factor f 2. Biom Biostat Int J. 2018;7(5):397-403. DOI: 10.15406/bbij.2018.07.00237

Download PDF

Abstract

Parametric and non-parametric bootstrap methods are used to investigate the statistical properties of the dissolution similarity factor f 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadAgadaWgaa WcbaGaaGOmaaqabaaaaa@38E1@ . The main objective of this study is to compare the results obtained by these two methods. We estimate characteristics of the sampling distribution of f ^ 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadAgagaqcam aaBaaaleaacaaIYaaabeaaaaa@38F1@  statistic under these methods with various bootstrap sample sizes using Monte Carlo simulation. A number of bootstrap confidence interval (CI) construction techniques are used to determine a 90 % CI for the true value of f 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadAgadaWgaa WcbaGaaGOmaaqabaaaaa@38E1@ under both parametric and non-parametric schemes. The bootstrap sampling distributions of f ^ 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadAgagaqcam aaBaaaleaacaaIYaaabeaaaaa@38F1@  under both schemes are found to be approximately symmetrical with a non-zero excess of kurtosis. Non-parametric bootstrap confidence intervals for f 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadAgadaWgaa WcbaGaaGOmaaqabaaaaa@38E1@  perform better than those obtained from parametric methods. The Bias corrected (BC) and accelerated bootstrap percentile (BCa) confidence interval method produce more precise two-sided confidence intervals for f 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadAgadaWgaa WcbaGaaGOmaaqabaaaaa@38E1@ compared to other methods

Keywords: dissolution profiles, bootstrapping, confidence interval, bias-corrected and accelerated bootstrap percentile confidence interval

Introduction

In pharmaceutical studies for solid and oral drugs, it is important to compare a test drug to a reference drug using average dissolution rates over time. The purpose of dissolution testing is to develop a new formulation, to ensure quality control, and to assess stability and reproducibility of the immediately released solid oral drug.1‒3 Assessment of dissolution profiles for two drugs, in vitro, provides the waiver for in-vivo assessment.

The United States Food and Drug Administration (FDA) requires similarity tests for the dissolution profiles of two drugs under consideration when there are post drug-approval changes. Such changes include change of manufacturing sites, change in formulations, and change in component and composition. Despite the post-approval changes, two drugs are similar with respect to their dissolution rates if the test (post-approval) has the same (equivalent) dissolution performance as the reference (pre-change).

In order to assess drug dissolution profiles, both model-dependent and model-independent methods are used. In a model-dependent approach, an appropriate mathematical model is selected to describe the dissolution profiles of the two drugs. The model is then fit to the data and confidence intervals for the model parameters are constructed. These confidence intervals are then compared with the specified similarity region. Commonly used model-dependent methods to fit the dissolution profiles include Gompertz,4 Logistic,5 Weibull,6 probit and sigmoid models.7,8 Model-dependent methods have some limitations. For example, selecting an appropriate model, and interpreting its parameters are difficult when the dissolution profiles for the two drugs follow different models.

To overcome the limitations of model-dependent approaches, model-independent approaches such as difference factor f 1 MathType@MTEF@5@5@+= feaagKart1ev2aqatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaabaaaaaaaaape GaamOza8aadaWgaaWcbaWdbiaaigdaa8aabeaaaaa@392F@ , similarity factor f 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaabaaaaaaaaape GaamOza8aadaWgaaWcbaWdbiaaikdaa8aabeaaaaa@392F@ , analysis of variance, split plot analysis, repeated measure analysis, Hotelling T 2 MathType@MTEF@5@5@+= feaagKart1ev2aqatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaabaaaaaaaaape Gaamiva8aadaahaaWcbeqaa8qacaaIYaaaaaaa@3910@ , principal component analysis1 and first order autoregressive time series analysis are used. Among these methods, analysis of variance, split plot analysis assume that dissolution data are independent over time. These two methods are not appropriate in many cases as data are not independent. As an alternative, Tsong et al.,9 proposed Hotelling T 2 MathType@MTEF@5@5@+= feaagKart1ev2aqatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaabaaaaaaaaape Gaamiva8aadaahaaWcbeqaa8qacaaIYaaaaaaa@3910@  statistic to construct a 90% confidence region for the difference in dissolution means of two batches of the reference product at two time points. This confidence region is then compared with a pre-specified similarity region.

Of all the model-independent approaches, the US FDA recommends only f 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaabaaaaaaaaape GaamOza8aadaWgaaWcbaWdbiaaikdaa8aabeaaaaa@392F@ 10 to study similarity between two drug dissolution profiles under consideration. Although this similarity factor f 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaabaaaaaaaaape GaamOza8aadaWgaaWcbaWdbiaaikdaa8aabeaaaaa@392F@  is used to assess global similarity of dissolution profiles, and it does not require any assumption regarding data generating process, using point estimate of f 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaabaaaaaaaaape GaamOza8aadaWgaaWcbaWdbiaaikdaa8aabeaaaaa@392F@ in comparing two drug dissolution profiles is not appropriate if there is substantial variation from batch to batch. In this case, it is necessary to construct the confidence intervals for f 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaabaaaaaaaaape GaamOza8aadaWgaaWcbaWdbiaaikdaa8aabeaaaaa@392F@ . Construction of confidence interval for f 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaabaaaaaaaaape GaamOza8aadaWgaaWcbaWdbiaaikdaa8aabeaaaaa@392F@  depends on the standard error of its estimator f ^ 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadAgagaqcam aaBaaaleaacaaIYaaabeaaaaa@38F1@ . Since there is no closed form formula for the standard error of f ^ 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadAgagaqcam aaBaaaleaacaaIYaaabeaaaaa@38F1@ , and hard to derive analytically, we approximate the standard error of f ^ 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadAgagaqcam aaBaaaleaacaaIYaaabeaaaaa@38F1@  by deriving sampling distribution of f ^ 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadAgagaqcam aaBaaaleaacaaIYaaabeaaaaa@38F1@  using parametric and non-parametric bootstrap methods. Then the approximated standard error of f ^ 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadAgagaqcam aaBaaaleaacaaIYaaabeaaaaa@38F1@  is used to construct bootstrap confidence intervals for f 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaabaaaaaaaaape GaamOza8aadaWgaaWcbaWdbiaaikdaa8aabeaaaaa@392F@ .

The organization of this paper is as follows: in section 2 we present basic characteristics of drug dissolution data used in our study and chi-square plot for assessing the normality of the underlying population of the data. Section 3 discusses the statistical framework used in dissolution testing and gives an outline how two drugs are considered to be similar in terms of dissolved drug ingredients into the media. In section 4, bootstrap methods are briefly discussed and related confidence intervals for the true value of f ^ 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadAgagaqcam aaBaaaleaacaaIYaaabeaaaaa@38F1@  are presented. Section 5 discusses the results of our study and section 6 concludes the paper.

Dissolution data

We consider the standard dissolution data discussed by Chow & Liu,11 Tsong12,13 to assess the various characteristics of f ^ 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadAgagaqcam aaBaaaleaacaaIYaaabeaaaaa@38F1@ . The Summary measures of the reference and the test drug dissolution data are presented in Table 1.

Time (Hour)

1

2

3

4

6

8

10

Test Drug

-

-

-

-

-

-

-

Mean

36.5

50.08

62.17

67.92

79.33

86.42

92

St. Deviation

1.38

2.27

1.47

3

2.53

3.73

2.41

Range

5

9

5

12

8

12

8

Reference Drug

-

-

-

-

-

-

-

Mean

45.08

54

62.5

67.08

74.75

80.25

85.33

St. Deviation

3.2

3.41

3.32

3.75

3.93

4.2

4.5

Range

12

11

14

15

15

16

17

Mean Difference

-8.58

-3.92

-0.33

0.83

4.58

6.17

6.67

Table 1 Summary measures of test and reference drug dissolution data

The observed difference between mean dissolution rate factors, f 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaabaaaaaaaaape GaamOza8aadaWgaaWcbaWdbiaaikdaa8aabeaaaaa@392F@ for the test and the reference drug at different time points are less than 10 percent. The standard deviation of the dissolution rate factor at different time points for the test and the reference drug are also less than 10 percent. The mean differences between the two drugs are wider at the starting time points than in the mid-time points. Figure 1 shows dissolution profiles for the test and the reference drugs.

Figure 1 Dissolution profiles for the test and the reference drugs

In order to perform parametric bootstrapping using the above data, we need to know the parametric form for the distribution of the population from which these sample dissolution factors are drawn. In particular we check if the sample dissolution factors are drawn from multivariate normal distribution. To check normality, we examine the underlying distribution of the data using a chi-square plot and a normality goodness of fit test. Because the observations from the same tablets across time are related and the observations across the tablets at a fixed time point are independent, the dissolution data used in this article are considered to be a realization of multivariate observations. To check whether the dissolution data we consider for our study come from the multivariate normal distribution, we calculate statistical distance measures and use them to construct a chi-square plot under the normality assumption (Figure 2).

Figure 2 Left panel: Chi-square plot for the test drug. Right panel: chi-square plot for the reference drug

Since the points in plots are not on a straight line, we say that the data do not follow multivariate normal distributions. We also use the formal correlation test to measure the straightness of the Q-Q plot. The values of the correlation coefficient for the Q-Q plot for the test and the reference drug dissolution data are 0.95 and 0.91 respectively. At 5% level of significance the tabulated value of the correlation coefficient for sample size of n=12 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqiVu0Je9sqqrpepC0xbbL8F4rqqrFfpeea0xe9Lq=Jc9 vqaqpepm0xbba9pwe9Q8fs0=yqaqpepae9pg0FirpepeKkFr0xfr=x fr=xb9adbaqaaeGaciGaaiaabeqaamaabaabaaGcbaqcLbsacaWGUb Gaeyypa0JaaGymaiaaikdaaaa@39F5@  is 0.9298. For the test drug dissolution data, the normality assumption is reasonable but for the reference drug, normality assumption is off slightly.

Statistical methods for drug dissolution

Let y ijt MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadMhammaaBa aaleaajugWaiaadMgacaWGQbGaamiDaaWcbeaaaaa@3C53@  be the percentage of drug dissolved in a media at time point t from the tablet i for drug j. Then the statistical model for the drug dissolution percentage can be written as y ijk = μ jt + ε ijt MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadMhadaWgaa WcbaGaamyAaiaadQgacaWGRbaabeaakiabg2da9iabeY7aTnaaBaaa leaacaWGQbGaamiDaaqabaGccqGHRaWkcqaH1oqzlmaaBaaabaqcLb macaWGPbGaamOAaiaadshaaSqabaaaaa@46AD@ , t=1,2,...,T MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadshacqGH9a qpcaaIXaGaaiilaiaaikdacaGGSaGaaiOlaiaac6cacaGGUaGaaiil aiaadsfaaaa@3F83@ ; i=1,2,...,n MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadMgacqGH9a qpcaaIXaGaaiilaiaaikdacaGGSaGaaiOlaiaac6cacaGGUaGaaiil aiaad6gaaaa@3F92@ ; j=1,2. MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadQgacqGH9a qpcaaIXaGaaiilaiaaikdacaGGUaaaaa@3BDC@  Here μ jt MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiabeY7aTnaaBa aaleaacaWGQbGaamiDaaqabaaaaa@3AD8@ is the population mean over tablets at time t for drug j MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadQgaaaa@37FD@  and ε ijt MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiabew7aLnaaBa aaleaacaWGPbGaamOAaiaadshaaeqaaaaa@3BB7@  has mean 0. Since the dissolution percentage is measured over time from the same tablet of the drug, the measurements are dependent. However, it is reasonable to assume that the n MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaad6gaaaa@3801@  vectors y 1j , y 2j ,..., y nj ,j=1,2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaahMhadaWgaa WcbaGaaGymaiaadQgaaeqaaOGaaiilaiaahMhadaWgaaWcbaGaaGOm aiaadQgaaeqaaOGaaiilaiaac6cacaGGUaGaaiOlaiaacYcacaWH5b WaaSbaaSqaaiaad6gacaWGQbaabeaakiaacYcacaWGQbGaeyypa0Ja aGymaiaacYcacaaIYaaaaa@48DF@  are independent, as these are replications across tablets in the population. The dissolution profiles (Figure 1) of the test drug and the reference drug are considered to be similar if and only if the population means vector for the test drug is in some neighborhood of the population mean vector for the reference drug. A rectangular similarity measure recommended and required by FDA is used to assess if two drugs are similar. This similarity measure is | μ 1t μ 2t |Δ MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaamaaemaabaGaeq iVd02aaSbaaSqaaiaaigdacaWG0baabeaakiabgkHiTiabeY7aTnaa BaaaleaacaaIYaGaamiDaaqabaaakiaawEa7caGLiWoacqGHKjYOcq qHuoaraaa@4579@ , where Δ MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiabfs5aebaa@3874@ a specified number is. Generally FDA recommends that the specified number is 10 for all time point. A similarity measure f 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadAgadaWgaa WcbaGaaGOmaaqabaaaaa@38E1@ , based on the rectangular measure and recommended by FDA, is discussed in Section 3.1.

Similarity factor f 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadAgadaWgaa WcbaGaaGOmaaqabaaaaa@38E1@

Moore et al.,10 developed a similarity factor, f 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadAgadaWgaa WcbaGaaGOmaaqabaaaaa@38E1@ , for testing dissolution profiles of a test and a reference drug.

The f 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadAgadaWgaa WcbaGaaGOmaaqabaaaaa@38E1@  similarity factor is defined as

f 2 =50log{ 100 ( 1+ D 2 ) 1 2 } MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadAgadaWgaa WcbaGaaGOmaaqabaGccqGH9aqpcaaI1aGaaGimaiGacYgacaGGVbGa ai4zamaacmaabaGaaGymaiaaicdacaaIWaWaaeWaaeaacaaIXaGaey 4kaSIaamiramaaCaaaleqabaGaaGOmaaaaaOGaayjkaiaawMcaamaa CaaaleqabaGaeyOeI0YaaSaaaeaacaaIXaaabaGaaGOmaaaaaaaaki aawUhacaGL9baaaaa@4A27@ ,

where D 2 = 1 T t=1 T ( μ 1t μ 2t ) 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadseadaahaa WcbeqaaiaaikdaaaGccqGH9aqpdaWcaaqaaiaaigdaaeaacaWGubaa amaaqahabaWaaeWaaeaacqaH8oqBdaWgaaWcbaGaaGymaiaadshaae qaaOGaeyOeI0IaeqiVd02aaSbaaSqaaiaaikdacaWG0baabeaaaOGa ayjkaiaawMcaamaaCaaaleqabaGaaGOmaaaaaeaacaWG0bGaeyypa0 JaaGymaaqaaiaadsfaa0GaeyyeIuoaaaa@4BDE@ and μ jt MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiabeY7aTnaaBa aaleaacaWGQbGaamiDaaqabaaaaa@3AD8@ is population mean dissolution rates over time t( t=1,2,...,T ) MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadshadaqada qaaiaadshacqGH9aqpcaaIXaGaaiilaiaaikdacaGGSaGaaiOlaiaa c6cacaGGUaGaaiilaiaadsfaaiaawIcacaGLPaaaaaa@4205@ and for the jth MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadQgacaqG0b GaaeiAaaaa@39DF@ drug ( j=1,2 ) MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaamaabmaabaGaam OAaiabg2da9iaaigdacaGGSaGaaGOmaaGaayjkaiaawMcaaaaa@3CB3@ . D 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadseadaahaa Wcbeqaaiaaikdaaaaaaa@38C0@ is a squared distance from the population mean vector of the test drug to the population mean vector of the reference drug. Since dissolution measurements are expressed as percent, D 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadseadaahaa Wcbeqaaiaaikdaaaaaaa@38C0@ ranges from 0 to 100 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaaigdacaaIWa GaaGimamaaCaaaleqabaGaaGOmaaaaaaa@3A26@ .

The similarity factor f 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadAgadaWgaa WcbaGaaGOmaaqabaaaaa@38E1@  is a monotone decreasing function of D 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadseadaahaa Wcbeqaaiaaikdaaaaaaa@38C0@  with a maximum of 100 when D 2 =0 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadseadaahaa WcbeqaaiaaikdaaaGccqGH9aqpcaaIWaaaaa@3A8A@  (two dissolution profiles are the same), and a minimum of 0, when D 2 = 100 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadseadaahaa WcbeqaaiaaikdaaaGccqGH9aqpcaaIXaGaaGimaiaaicdadaahaaWc beqaaiaaikdaaaaaaa@3CE8@ .

A value of f 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadAgadaWgaa WcbaGaaGOmaaqabaaaaa@38E1@  in the range of 50 to 100 ensures the similarity or equivalence of two dissolution profiles. When the rectangular similarity measure (adopted by FDA) is | μ 2t μ 1t |10 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaamaaemaabaGaeq iVd02aaSbaaSqaaiaaikdacaWG0baabeaakiabgkHiTiabeY7aTnaa BaaaleaacaaIXaGaamiDaaqabaaakiaawEa7caGLiWoacqGHKjYOca aIXaGaaGimaaaa@4588@  for all time points, then f 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadAgadaWgaa WcbaGaaGOmaaqabaaaaa@38E1@  is very close to 50. So the similarity region in the range of 50 to 100 indicates the similarity of two drugs.

This similarity factor works well when the following conditions are met: (i) there is a minimum of three time points, (ii) there are 12 individual values for each time point for each formulation, (iii) no more than one mean value is greater than 85% dissolved for each formulation, and (iv) the standard deviation of the mean of any product is less than 10% from the second to last time points.

Bootstrap methods

Bootstrapping14,15 is a computer-intensive approach to statistical inference. It is based on the sampling distribution of a statistic obtained by resampling from the data with replacement. When it is hard to derive the exact sampling distribution of certain statistics and their characteristics, bootstrap methods are used to approximate them. The characteristics include standard error, bias, skewness, critical values, mean squared error, and others. To derive an exact sampling distribution of a statistic of interest, the underlying population distribution from which sample is drawn has to be known. Sometimes even though the underlying distribution is known, derivation of the exact sampling distribution for certain statistic is not possible or is very complex. In such case, bootstrap methods allow estimating or approximating the sampling distributions of these statistics. The bootstrap approach does not require knowledge of the data generating process but uses the sample information only. The idea behind bootstrapping is that the use of sample information as a “proxy population”. One takes samples with replacement from the original sample and calculates the statistic of interest repeatedly. This leads to a bootstrap sampling distribution. This sampling distribution is used to measure the estimator’s accuracy and helps to set approximate confidence intervals for certain population parameters.

We use two types of bootstrap methods, parametric and non-parametric to determine the sampling distribution of the statistic f ^ 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadAgagaqcam aaBaaaleaacaaIYaaabeaaaaa@38F1@ and its characteristics. Using both techniques we construct 90% confidence intervals for f 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadAgadaWgaa WcbaGaaGOmaaqabaaaaa@38E1@ . We briefly describe both methods as follows.

Let X 1 , X 2 ,..., X n MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadIfadaWgaa WcbaGaaGymaaqabaGccaGGSaGaamiwamaaBaaaleaacaaIYaaabeaa kiaacYcacaGGUaGaaiOlaiaac6cacaGGSaGaamiwamaaBaaaleaaca WGUbaabeaaaaa@40CD@ be independent and identically distributed random variables from an unknown distribution F ^ n MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadAeagaqcam aaBaaaleaacaWGUbaabeaaaaa@3908@ . F MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadAeaaaa@37D9@  is estimated using the empirical distribution F ^ n MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadAeagaqcam aaBaaaleaacaWGUbaabeaaaaa@3908@ . Repeated samples are taken from the estimated empirical distribution F ^ n MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadAeagaqcam aaBaaaleaacaWGUbaabeaaaaa@3908@ . Then the statistic of interest is calculated using each bootstrap samples, giving a set of bootstrap values for the desired statistic. Using the bootstrap values of the statistic, the estimated distribution function and its properties are calculated. This approach is called non-parametric bootstrapping.

The parametric bootstrap, on the other hand, assumes that F MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadAeaaaa@37D9@ is known except for its parameters. F MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadAeaaaa@37D9@ is approximated by estimating the parameters with the sample observations. Then from the approximated distribution F ^ n MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadAeagaqcam aaBaaaleaacaWGUbaabeaaaaa@3908@ , repeated samples are taken. The values of the statistic of interest are calculated using these bootstrap samples. These bootstrap values of the statistic are used to derive the desired measures. Under both schemes, the distribution of f ^ 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadAgagaqcam aaBaaaleaacaaIYaaabeaaaaa@38F1@  can be estimated by using the bootstrap with the Monte Carlo approximation as follows.

F(x)= 1 B b=1 B I{ f 2 * x }, MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadAeacaGGOa GaamiEaiaacMcacqGH9aqpdaWcaaqaaiaaigdaaeaacaWGcbaaamaa qahabaGaamysamaacmaabaGaamOzamaaDaaaleaacaaIYaaabaGaai OkaaaakiabgsMiJkaadIhaaiaawUhacaGL9baacaGGSaaaleaacaWG IbGaeyypa0JaaGymaaqaaiaadkeaa0GaeyyeIuoaaaa@4B65@

where f 2 * MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadAgadaqhaa WcbaGaaGOmaaqaaiaacQcaaaaaaa@3990@ is the value of the f 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadAgadaWgaa WcbaGaaGOmaaqabaaaaa@38E1@  based on the bootstrap sample and F(x) MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadAeacaGGOa GaamiEaiaacMcaaaa@3A2F@ is a bootstrap estimator of the distribution function of f ^ 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadAgagaqcam aaBaaaleaacaaIYaaabeaaaaa@38F1@ based on the data X 1 , X 2, ..., X n MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadIfadaWgaa WcbaGaaGymaaqabaGccaGGSaGaamiwamaaBaaaleaacaaIYaGaaiil aaqabaGccaGGUaGaaiOlaiaac6cacaGGSaGaamiwamaaBaaaleaaca WGUbaabeaaaaa@40CD@ . The bootstrap histogram for { f 2 * ,b=1,2,...,B } MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaamaacmaabaGaam OzamaaDaaaleaacaaIYaaabaGaaiOkaaaakiaacYcacaWGIbGaeyyp a0JaaGymaiaacYcacaaIYaGaaiilaiaac6cacaGGUaGaaiOlaiaacY cacaWGcbaacaGL7bGaayzFaaaaaa@44CC@ can be used to estimate the density of f ^ 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadAgagaqcam aaBaaaleaacaaIYaaabeaaaaa@38F1@ . The expected value, variance, skewness, kurtosis, and bias of the bootstrap sampling distribution of f ^ 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadAgagaqcam aaBaaaleaacaaIYaaabeaaaaa@38F1@  are estimated from F(x) MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadAeacaGGOa GaamiEaiaacMcaaaa@3A2F@ . In order compute them, we first take B independent samples { X 1 *,b , X 2 *,b , X 3 *,b ,..., X n *,b } MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaamaacmaabaGaam iwamaaDaaaleaacaaIXaaabaGaaiOkaiaacYcacaWGIbaaaOGaaiil aiaadIfadaqhaaWcbaGaaGOmaaqaaiaacQcacaGGSaGaamOyaaaaki aacYcacaWGybWaa0baaSqaaiaaiodaaeaacaGGQaGaaiilaiaadkga aaGccaGGSaGaaiOlaiaac6cacaGGUaGaaiilaiaadIfadaqhaaWcba GaamOBaaqaaiaacQcacaGGSaGaamOyaaaaaOGaay5Eaiaaw2haaaaa @4EA0@ , b=1,2,...,B MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadkgacqGH9a qpcaaIXaGaaiilaiaaikdacaGGSaGaaiOlaiaac6cacaGGUaGaaiil aiaadkeaaaa@3F5F@  and approximate them by

f ^ ¯ 2 = 1 B b=1 B f 2,b * MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadAgagaqcga qeamaaBaaaleaacaaIYaaabeaakiabg2da9maalaaabaGaaGymaaqa aiaadkeaaaWaaabCaeaacaWGMbWaa0baaSqaaiaaikdacaGGSaGaam OyaaqaaiaacQcaaaaabaGaamOyaiabg2da9iaaigdaaeaacaWGcbaa niabggHiLdaaaa@4569@ , υ boot ( B ) = 1 B b=1 B ( f 2,b * f ^ ¯ 2 ) 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiabew8a1naaDa aaleaacaqGIbGaae4Baiaab+gacaqG0baabaWaaeWaaeaacaWGcbaa caGLOaGaayzkaaaaaOGaeyypa0ZaaSaaaeaacaaIXaaabaGaamOqaa aadaaeWbqaamaabmaabaGaamOzamaaDaaaleaacaaIYaGaaiilaiaa dkgaaeaacaGGQaaaaOGaeyOeI0IabmOzayaajyaaraWaaSbaaSqaai aaikdaaeqaaaGccaGLOaGaayzkaaaaleaacaWGIbGaeyypa0JaaGym aaqaaiaadkeaa0GaeyyeIuoakmaaCaaaleqabaGaaGOmaaaaaaa@50F5@ ,

s k boot ( B ) = 1 B b=1 B ( f 2,b * f ^ ¯ 2 ) 3 [ υ boot ( B ) ] 3 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadohacaWGRb Waa0baaSqaaiaabkgacaqGVbGaae4Baiaabshaaeaadaqadaqaaiaa dkeaaiaawIcacaGLPaaaaaGccqGH9aqpdaWcaaqaamaalaaabaGaaG ymaaqaaiaadkeaaaWaaabCaeaadaqadaqaaiaadAgadaqhaaWcbaGa aGOmaiaacYcacaWGIbaabaGaaiOkaaaakiabgkHiTiqadAgagaqcga qeamaaBaaaleaacaaIYaaabeaaaOGaayjkaiaawMcaaaWcbaGaamOy aiabg2da9iaaigdaaeaacaWGcbaaniabggHiLdGcdaahaaWcbeqaai aaiodaaaaakeaadaWadaqaaiabew8a1naaDaaaleaacaqGIbGaae4B aiaab+gacaqG0baabaWaaeWaaeaacaWGcbaacaGLOaGaayzkaaaaaa GccaGLBbGaayzxaaWaaWbaaSqabeaadaWcaaqaaiaaiodaaeaacaaI Yaaaaaaaaaaaaa@5CE7@ , k boot ( B ) = 1 B b=1 B ( f 2,b * f ^ ¯ 2 ) 4 [ υ boot ( B ) ] 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadUgadaqhaa WcbaGaaeOyaiaab+gacaqGVbGaaeiDaaqaamaabmaabaGaamOqaaGa ayjkaiaawMcaaaaakiabg2da9maalaaabaWaaSaaaeaacaaIXaaaba GaamOqaaaadaaeWbqaamaabmaabaGaamOzamaaDaaaleaacaaIYaGa aiilaiaadkgaaeaacaGGQaaaaOGaeyOeI0IabmOzayaajyaaraWaaS baaSqaaiaaikdaaeqaaaGccaGLOaGaayzkaaaaleaacaWGIbGaeyyp a0JaaGymaaqaaiaadkeaa0GaeyyeIuoakmaaCaaaleqabaGaaGinaa aaaOqaamaadmaabaGaeqyXdu3aa0baaSqaaiaabkgacaqGVbGaae4B aiaabshaaeaadaqadaqaaiaadkeaaiaawIcacaGLPaaaaaaakiaawU facaGLDbaadaahaaWcbeqaaiaaikdaaaaaaaaa@5B23@

and b Boot ( B ) = f ^ ¯ 2 f ^ 2 ; MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadkgadaqhaa WcbaGaamOqaiaad+gacaWGVbGaamiDaaqaamaabmaabaGaamOqaaGa ayjkaiaawMcaaaaakiabg2da9iqadAgagaqcgaqeaSWaaSbaaeaaju gWaiaaikdaaSqabaGccqGHsislceWGMbGbaKaadaWgaaWcbaGaaGOm aaqabaGccaGG7aaaaa@4600@

here f ^ ¯ 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadAgagaqcga qeamaaBaaaleaacaaIYaaabeaaaaa@3908@ , υ boot ( B ) MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiabew8a1naaDa aaleaacaqGIbGaae4Baiaab+gacaqG0baabaWaaeWaaeaacaWGcbaa caGLOaGaayzkaaaaaaaa@3F12@ , s k boot ( B ) MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadohacaWGRb Waa0baaSqaaiaabkgacaqGVbGaae4Baiaabshaaeaadaqadaqaaiaa dkeaaiaawIcacaGLPaaaaaaaaa@3F33@ , k boot ( B ) MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadUgadaqhaa WcbaGaaeOyaiaab+gacaqGVbGaaeiDaaqaamaabmaabaGaamOqaaGa ayjkaiaawMcaaaaaaaa@3E3B@ , and b Boot ( B ) MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadkgadaqhaa WcbaGaamOqaiaad+gacaWGVbGaamiDaaqaamaabmaabaGaamOqaaGa ayjkaiaawMcaaaaaaaa@3E1A@  are Monte Carlo bootstrap estimator for mean, variance, skewness, kurtosis and bias of the sampling distribution of f ^ MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadAgagaqcaa aa@3809@  respectively. In section 4, we construct bootstrap confidence intervals for f 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadAgadaWgaa WcbaGaaGOmaaqabaaaaa@38E1@  using a number of available bootstrap confidence interval methods.

Bootstrap confidence intervals

An observed value of f ^ 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadAgagaqcam aaBaaaleaacaaIYaaabeaaaaa@38F1@  is used to assess whether two drugs (test and reference) are similar or not with respect to their dissolution profiles. This value is compared with the specifications given by the FDA in order to decide if the two drugs are similar. However, due to sampling variation, it is not reasonable to assess the dissolution similarity of two drugs by directly comparing f ^ 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadAgagaqcam aaBaaaleaacaaIYaaabeaaaaa@38F1@ with the specification limits. Rather one can make a decision of dissolution similarity by constructing a 90% confidence interval for the population parameter f 2 =E( f ^ 2 ) MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadAgadaWgaa WcbaGaaGOmaaqabaGccqGH9aqpcaWGfbGaaiikaiqadAgagaqcamaa BaaaleaacaaIYaaabeaakiaacMcaaaa@3E01@ . In this section we use parametric and non-parametric bootstrap methods discussed in subsection 3.2 to construct the confidence intervals for f 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadAgadaWgaa WcbaGaaGOmaaqabaaaaa@38E1@ .

A detailed discussion on different types of bootstrap confidence intervals can be found in Chernick,16 Davison,17 DiCiccio,18 Efron.19 Here we review different types of bootstrap procedures used to construct confidence intervals for the parameter of interest briefly. For notational convenience, we denote the similarity parameter f 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadAgadaWgaa WcbaGaaGOmaaqabaaaaa@38E1@ as θ MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiabeI7aXbaa@38C4@  and θ ^ MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqbeI7aXzaaja aaaa@38D4@  as its estimate.

The standard bootstrap confidence interval is given by

[ θ ^ z ( α 2 ) s e ^ ( θ ^ * ) , θ ^ + z ( 1 α 2 ) s e ^ ( θ ^ * ) ] MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaamaadmaabaGafq iUdeNbaKaacqGHsislcaWG6bWaaWbaaSqabeaadaqadaqaamaalaaa baGaeqySdegabaGaaGOmaaaaaiaawIcacaGLPaaaaaGccaWGZbWaae caaeaacaWGLbaacaGLcmaadaWgaaWcbaGaaiikaiqbeI7aXzaajaWa aWbaaWqabeaacaGGQaaaaSGaaiykaaqabaGccaGGSaGafqiUdeNbaK aacqGHRaWkcaWG6bWaaWbaaSqabeaadaqadaqaaiaaigdacqGHsisl daWcaaqaaiabeg7aHbqaaiaaikdaaaaacaGLOaGaayzkaaaaaOGaam 4CamaaHaaabaGaamyzaaGaayPadaWaaSbaaSqaaiaacIcacuaH4oqC gaqcamaaCaaameqabaGaaiOkaaaaliaacMcaaeqaaaGccaGLBbGaay zxaaaaaa@58C7@

where s e ^ ( θ ^ * ) MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadohadaqiaa qaaiaadwgaaiaawkWaamaaBaaaleaacaGGOaGafqiUdeNbaKaadaah aaadbeqaaiaacQcaaaWccaGGPaaabeaaaaa@3DE4@  is the bootstrap standard error of the estimator θ ^ MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqbeI7aXzaaja aaaa@38D4@ , and z ( α 2 ) MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadQhadaahaa WcbeqaamaabmaabaWaaSaaaeaacqaHXoqyaeaacaaIYaaaaaGaayjk aiaawMcaaaaaaaa@3C2E@  is the 100. α 2 th MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaaigdacaaIWa GaaGimaiaac6cadaWcaaqaaiabeg7aHbqaaiaaikdaaaGaamiDaiaa dIgaaaa@3E40@  quantile of the standard normal distribution. In percentile interval method of bootstrapping, B MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadkeaaaa@37D5@ bootstrap estimates θ ^ b * MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqbeI7aXzaaja Waa0baaSqaaiaadkgaaeaacaGGQaaaaaaa@3A96@ ( b=1,2,...,B ) MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaamaabmaabaGaam Oyaiabg2da9iaaigdacaGGSaGaaGOmaiaacYcacaGGUaGaaiOlaiaa c6cacaGGSaGaamOqaaGaayjkaiaawMcaaaaa@40E8@ are generated. Then these bootstrap estimates are arranged in ascending order. If we denote F( θ ^ * ) MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadAeadaqada qaaiqbeI7aXzaajaWaaWbaaSqabeaacaGGQaaaaaGccaGLOaGaayzk aaaaaa@3C0D@  as the cumulative distribution function of θ ^ * MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqbeI7aXzaaja WaaWbaaSqabeaacaGGQaaaaaaa@39AF@ , then a 90% percentile interval is defined by

( θ ^ b *( α 2 ) , θ ^ b *( 1 α 2 ) )=[ F ^ 1 ( α 2 ), F ^ 1 ( 1 α 2 ) ] MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaamaabmaabaGafq iUdeNbaKaadaqhaaWcbaGaamOyaaqaaiaacQcadaqadaqaamaalaaa baGaeqySdegabaGaaGOmaaaaaiaawIcacaGLPaaaaaGccaGGSaGafq iUdeNbaKaadaqhaaWcbaGaamOyaaqaaiaacQcadaqadaqaaiaaigda cqGHsisldaWcaaqaaiabeg7aHbqaaiaaikdaaaaacaGLOaGaayzkaa aaaaGccaGLOaGaayzkaaGaeyypa0ZaamWaaeaaceWGgbGbaKaadaah aaWcbeqaaiabgkHiTiaaigdaaaGcdaqadaqaamaalaaabaGaeqySde gabaGaaGOmaaaaaiaawIcacaGLPaaacaGGSaGabmOrayaajaWaaWba aSqabeaacqGHsislcaaIXaaaaOWaaeWaaeaacaaIXaGaeyOeI0YaaS aaaeaacqaHXoqyaeaacaaIYaaaaaGaayjkaiaawMcaaaGaay5waiaa w2faaaaa@5CA7@

where α=5% MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiabeg7aHjabg2 da9iaaiwdacaGGLaaaaa@3B1B@  and θ ^ *( α 2 ) MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqbeI7aXzaaja WaaWbaaSqabeaacaGGQaWaaeWaaeaadaWcaaqaaiabeg7aHbqaaiaa ikdaaaaacaGLOaGaayzkaaaaaaaa@3DA3@  indicates the 100. α 2 th MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaaigdacaaIWa GaaGimaiaac6cadaWcaaqaaiabeg7aHbqaaiaaikdaaaGaamiDaiaa dIgaaaa@3E40@  percentile of B MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadkeaaaa@37D5@  bootstrap replications.

Although the computation is straightforward, this method does not work well when the sampling distribution of θ ^ MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqbeI7aXzaaja aaaa@38D4@  is skewed or θ ^ MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqbeI7aXzaaja aaaa@38D4@  is biased.20,21

In the presence of skewness and bias, the percentile method can be improved by an adjustment to the percentile method. This bias adjusted and corrected percentile interval is known as bias corrected percentile interval method (BC).22 In the bias-corrected method, the observed amount of difference between the median of the bootstrap estimate θ ^ * MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqbeI7aXzaaja WaaWbaaSqabeaacaGGQaaaaaaa@39AF@ and the observed estimate from the original sample is defined as bias. The bias-correction constant estimate, denoted by z ^ 0 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqiVu0Je9sqqrpepC0xbbL8F4rqqrFfpeea0xe9Lq=Jc9 vqaqpepm0xbba9pwe9Q8fs0=yqaqpepae9pg0FirpepeKkFr0xfr=x fr=xb9adbaqaaeGaciGaaiaabeqaamaabaabaaGcbaGabmOEayaaja WaaSbaaSqaaiaaicdaaeqaaaaa@37EB@ , is defined as

z ^ 0 = Φ 1 ( # θ ^ * < θ ^ B ) MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadQhagaqcam aaBaaaleaacaaIWaaabeaakiabg2da9iabfA6agnaaCaaaleqabaGa eyOeI0IaaGymaaaakmaabmaabaWaaSaaaeaacaGGJaGafqiUdeNbaK aadaahaaWcbeqaaiaacQcaaaGccqGH8aapcuaH4oqCgaqcaaqaaiaa dkeaaaaacaGLOaGaayzkaaaaaa@45E8@ ,

where Φ 1 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiabfA6agnaaCa aaleqabaGaeyOeI0IaaGymaaaaaaa@3A5D@  is the inverse function of a standard normal cumulative distribution function. Then, a 100( 1α ) MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaaigdacaaIWa GaaGimamaabmaabaGaaGymaiabgkHiTiabeg7aHbGaayjkaiaawMca aaaa@3E0D@ percent bias-corrected percentile confidence interval for θ MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiabeI7aXbaa@38C4@  is given by [ θ ^ *( α 1 ) , θ ^ *( α 2 ) ] MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaamaadmaabaGafq iUdeNbaKaadaahaaWcbeqaaiaacQcadaqadaqaaiabeg7aHnaaBaaa meaacaaIXaaabeaaaSGaayjkaiaawMcaaaaakiaacYcacuaH4oqCga qcamaaCaaaleqabaGaaiOkamaabmaabaGaeqySde2aaSbaaWqaaiaa ikdaaeqaaaWccaGLOaGaayzkaaaaaaGccaGLBbGaayzxaaaaaa@473D@ ,

where

α 1 =Φ( 2 z 0 + z ( α 2 ) ) MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiabeg7aHnaaBa aaleaacaaIXaaabeaakiabg2da9iabfA6agnaabmaabaGaaGOmaiaa dQhadaWgaaWcbaGaaGimaaqabaGccqGHRaWkcaWG6bWaaWbaaSqabe aadaqadaqaamaalaaabaGaeqySdegabaGaaGOmaaaaaiaawIcacaGL PaaaaaaakiaawIcacaGLPaaaaaa@465E@
α 2 =Φ( 2 z 0 + z ( 1 α 2 ) ) MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiabeg7aHnaaBa aaleaacaaIYaaabeaakiabg2da9iabfA6agnaabmaabaGaaGOmaiaa dQhadaWgaaWcbaGaaGimaaqabaGccqGHRaWkcaWG6bWaaWbaaSqabe aadaqadaqaaiaaigdacqGHsisldaWcaaqaaiabeg7aHbqaaiaaikda aaaacaGLOaGaayzkaaaaaaGccaGLOaGaayzkaaaaaa@4807@ .

Here Φ MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiabfA6agbaa@3888@  is the standard normal cumulative distribution function and z ( α 2 ) MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadQhadaahaa WcbeqaamaabmaabaWaaSaaaeaacqaHXoqyaeaacaaIYaaaaaGaayjk aiaawMcaaaaaaaa@3C2E@  is the 100 α 2 th MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaaigdacaaIWa GaaGimamaalaaabaGaeqySdegabaGaaGOmaaaacaqG0bGaaeiAaaaa @3D8A@  percentile point of the standard normal distribution. Although the bootstrap bias correction improves the bootstrap percentile method with taking the bias into account, this method does not work well in some cases.21

Efron22 introduced a further improved bootstrap method that corrects the bias due to the non-normality and also accelerates convergence to a solution. The method corrects the rate of change of the normalized standard error of θ ^ MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqbeI7aXzaaja aaaa@38D4@  relative to the true parameter θ MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiabeI7aXbaa@38C4@ . It takes into account the skewness in the distribution along with the bias of the estimator. This method is called bias-corrected and accelerated (BCa) percentile method. Chernick et al.,16 show that for small sample sizes BCa may not work as percentile method because the bias and acceleration constant must be estimated and the sample size is not large enough for asymptotic advantage of BCa to hold.

The BCa confidence interval for θ MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiabeI7aXbaa@38C4@  is

[ θ ^ *( α 1 * ) , θ ^ *( α 2 * ) ] MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaamaadmaabaGafq iUdeNbaKaadaahaaWcbeqaaiaacQcadaqadaqaaiabeg7aHnaaDaaa meaacaaIXaaabaGaaiOkaaaaaSGaayjkaiaawMcaaaaakiaacYcacu aH4oqCgaqcamaaCaaaleqabaGaaiOkamaabmaabaGaeqySde2aa0ba aWqaaiaaikdaaeaacaGGQaaaaaWccaGLOaGaayzkaaaaaaGccaGLBb Gaayzxaaaaaa@489B@ ,

where

α 1 * =Φ( z ^ 0 + z ^ 0 + z ( α 2 ) 1a( z ^ 0 + z ( α 2 ) ) ) MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiabeg7aHnaaDa aaleaacaaIXaaabaGaaiOkaaaakiabg2da9iabfA6agnaabmaabaGa bmOEayaajaWaaSbaaSqaaiaaicdaaeqaaOGaey4kaSYaaSaaaeaace WG6bGbaKaadaWgaaWcbaGaaGimaaqabaGccqGHRaWkcaWG6bWaaWba aSqabeaadaqadaqaamaalaaabaGaeqySdegabaGaaGOmaaaaaiaawI cacaGLPaaaaaaakeaacaaIXaGaeyOeI0IaamyyamaabmaabaGabmOE ayaajaWaaSbaaSqaaiaaicdaaeqaaOGaey4kaSIaamOEamaaCaaale qabaWaaeWaaeaadaWcaaqaaiabeg7aHbqaaiaaikdaaaaacaGLOaGa ayzkaaaaaaGccaGLOaGaayzkaaaaaaGaayjkaiaawMcaaaaa@5574@  and

α 2 * =Φ( z ^ 0 + z ^ 0 + z ( 1 α 2 ) 1a( z ^ 0 + z ( 1 α 2 ) ) ) MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiabeg7aHnaaDa aaleaacaaIYaaabaGaaiOkaaaakiabg2da9iabfA6agnaabmaabaGa bmOEayaajaWaaSbaaSqaaiaaicdaaeqaaOGaey4kaSYaaSaaaeaace WG6bGbaKaadaWgaaWcbaGaaGimaaqabaGccqGHRaWkcaWG6bWaaWba aSqabeaadaqadaqaaiaaigdacqGHsisldaWcaaqaaiabeg7aHbqaai aaikdaaaaacaGLOaGaayzkaaaaaaGcbaGaaGymaiabgkHiTiaadgga daqadaqaaiqadQhagaqcamaaBaaaleaacaaIWaaabeaakiabgUcaRi aadQhadaahaaWcbeqaamaabmaabaGaaGymaiabgkHiTmaalaaabaGa eqySdegabaGaaGOmaaaaaiaawIcacaGLPaaaaaaakiaawIcacaGLPa aaaaaacaGLOaGaayzkaaaaaa@58C5@ ,

where Φ MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiabfA6agbaa@3888@  is the standard normal cumulative distribution function and z ( . ) MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadQhadaahaa WcbeqaamaabmaabaGaaiOlaaGaayjkaiaawMcaaaaaaaa@3A75@  is the percentile of standard normal distribution.

The bias correction term z ^ 0 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadQhagaqcam aaBaaaleaacaaIWaaabeaaaaa@3903@  is calculated by z 0 = Φ 1 ( #( θ ^ * < θ ^ ) B ) MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadQhadaWgaa WcbaGaaGimaaqabaGccqGH9aqpcqqHMoGrdaahaaWcbeqaaiabgkHi TiaaigdaaaGcdaqadaqaamaalaaabaGaai4iamaabmaabaGafqiUde NbaKaadaahaaWcbeqaaiaacQcaaaGccqGH8aapcuaH4oqCgaqcaaGa ayjkaiaawMcaaaqaaiaadkeaaaaacaGLOaGaayzkaaaaaa@4761@ , and the acceleration constant a MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadggaaaa@37F4@ is :

a= i=1 n ( θ ^ ( . ) θ ^ ( i ) ) 3 6 [ i=1 n ( θ ^ ( . ) θ ^ ( i ) ) 2 ] 3 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadggacqGH9a qpdaWcaaqaamaaqahabaWaaeWaaeaacuaH4oqCgaqcamaaBaaaleaa daqadaqaaiaac6caaiaawIcacaGLPaaaaeqaaOGaeyOeI0IafqiUde NbaKaadaWgaaWcbaWaaeWaaeaacaWGPbaacaGLOaGaayzkaaaabeaa aOGaayjkaiaawMcaaaWcbaGaamyAaiabg2da9iaaigdaaeaacaWGUb aaniabggHiLdGcdaahaaWcbeqaaiaaiodaaaaakeaacaaI2aWaamWa aeaadaaeWbqaamaabmaabaGafqiUdeNbaKaadaWgaaWcbaWaaeWaae aacaGGUaaacaGLOaGaayzkaaaabeaakiabgkHiTiqbeI7aXzaajaWa aSbaaSqaamaabmaabaGaamyAaaGaayjkaiaawMcaaaqabaaakiaawI cacaGLPaaaaSqaaiaadMgacqGH9aqpcaaIXaaabaGaamOBaaqdcqGH ris5aOWaaWbaaSqabeaacaaIYaaaaaGccaGLBbGaayzxaaWaaWbaaS qabeaadaWcaaqaaiaaiodaaeaacaaIYaaaaaaaaaaaaa@6175@ ,

where θ ^ ( . ) MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqbeI7aXzaaja WaaSbaaSqaamaabmaabaGaaiOlaaGaayjkaiaawMcaaaqabaaaaa@3B3B@  and θ ^ ( i ) MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqbeI7aXzaaja WaaSbaaSqaamaabmaabaGaamyAaaGaayjkaiaawMcaaaqabaaaaa@3B77@ are the average and ith MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadMgacaqG0b GaaeiAaaaa@39DE@ jackknife estimate of the parameter.

The problem arising from the skewness in the sampling distribution of θ ^ MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqbeI7aXzaaja aaaa@38D4@  can also be handled by an alternative method called Bootstrap-t method.22,26,27 The bootstrap-t method is defined by the pivotal quantity t * = θ ^ * θ ^ s e θ ^ * MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadshadaahaa WcbeqaaiaacQcaaaGccqGH9aqpdaWcaaqaaiqbeI7aXzaajaWaaWba aSqabeaacaGGQaaaaOGaeyOeI0IafqiUdeNbaKaaaeaacaWGZbGaam yzamaaBaaaleaacuaH4oqCgaqcamaaCaaameqabaGaaiOkaaaaaSqa baaaaaaa@441B@ , where θ ^ * MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqbeI7aXzaaja WaaWbaaSqabeaacaGGQaaaaaaa@39AF@  and s e θ ^ * MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadohacaWGLb WaaSbaaSqaaiqbeI7aXzaajaWaaWbaaWqabeaacaGGQaaaaaWcbeaa aaa@3BC9@  are the bootstrap estimator and its standard error. Since the standard error of θ ^ MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqbeI7aXzaaja aaaa@38D4@  is not known, it is estimated by Monte Carlo simulation. However, this simulation requires nested bootstrapping. For each bootstrap sample, we calculate t * MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadshadaahaa WcbeqaaiaacQcaaaaaaa@38E2@ and these resulting t * MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadshadaahaa WcbeqaaiaacQcaaaaaaa@38E2@ ’s are placed in ascending order and select 100( α 2 )th  MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaaigdacaaIWa GaaGimamaabmaabaWaaSaaaeaacqaHXoqyaeaacaaIYaaaaaGaayjk aiaawMcaaiaabshacaqGObGaaeiiaaaa@3FB6@  and 100( 1 α 2 )th  MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaaigdacaaIWa GaaGimamaabmaabaGaaGymaiabgkHiTmaalaaabaGaeqySdegabaGa aGOmaaaaaiaawIcacaGLPaaacaqG0bGaaeiAaiaabccaaaa@415E@ percentile values of t * MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadshadaahaa WcbeqaaiaacQcaaaaaaa@38E2@ . Then, 100( 1α ) MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaaigdacaaIWa GaaGimamaabmaabaGaaGymaiabgkHiTiabeg7aHbGaayjkaiaawMca aaaa@3E0D@ percent bootstrap-t confidence for θ ^ MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqbeI7aXzaaja aaaa@38D4@  is

[ θ ^ t 1 α 2 * s e θ ^ * , θ ^ t α 2 * s e θ ^ * ] MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaamaadmaabaGafq iUdeNbaKaacqGHsislcaWG0bWaa0baaSqaaiaaigdacqGHsisldaWc aaqaaiabeg7aHbqaaiaaikdaaaaabaGaaiOkaaaakiaadohacaWGLb WaaSbaaSqaaiqbeI7aXzaajaWaaWbaaWqabeaacaGGQaaaaaWcbeaa kiaacYcacuaH4oqCgaqcaiabgkHiTiaadshadaqhaaWcbaWaaSaaae aacqaHXoqyaeaacaaIYaaaaaqaaiaacQcaaaGccaWGZbGaamyzamaa BaaaleaacuaH4oqCgaqcamaaCaaameqabaGaaiOkaaaaaSqabaaaki aawUfacaGLDbaaaaa@52DA@

where θ ^ MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqbeI7aXzaaja aaaa@38D4@  is estimate of the parameter θ MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiabeI7aXbaa@38C4@  from the original sample and s e θ ^ * MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadohacaWGLb WaaSbaaSqaaiqbeI7aXzaajaWaaWbaaWqabeaacaGGQaaaaaWcbeaa aaa@3BC9@ is the bootstrap standard error.

Results and discussion

Properties of the distribution of f ^ 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadAgagaqcam aaBaaaleaacaaIYaaabeaaaaa@38F1@

In this section, we examine the properties of the bootstrap sampling distribution of f ^ 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadAgagaqcam aaBaaaleaacaaIYaaabeaaaaa@38F1@  by using both non-parametric and parametric bootstrap sampling. To generate the bootstrap samples with non-parametric and parametric bootstrap methods discussed in subsection 3.3. the following algorithms are employed: (a) For Non-parametric bootstrapping: B(b=1,2,...,B) MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadkeacaGGOa GaamOyaiabg2da9iaaigdacaGGSaGaaGOmaiaacYcacaGGUaGaaiOl aiaac6cacaGGSaGaamOqaiaacMcaaaa@417F@ independent sample with replacement from the observed y 1j , y 2j ,..., y nj ,j=1,2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaahMhadaWgaa WcbaGaaGymaiaadQgaaeqaaOGaaiilaiaahMhadaWgaaWcbaGaaGOm aiaadQgaaeqaaOGaaiilaiaac6cacaGGUaGaaiOlaiaacYcacaWH5b WaaSbaaSqaaiaad6gacaWGQbaabeaakiaacYcacaWGQbGaeyypa0Ja aGymaiaacYcacaaIYaaaaa@48DF@ are drawn and for each bootstrap sample f ^ MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadAgagaqcaa aa@3809@ , which is the estimate of f 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadAgadaWgaa WcbaGaaGOmaaqabaaaaa@38E1@ defined in Subsection 3.1, is calucated; (b) For Parametric bootstrappping: B MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaceaaelGaamOqaa aa@389E@  independent samples are drawn from N T ( μ ^ j , ^ j ),j=1,2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaad6eadaWgaa WcbaGaamivaaqabaGccaGGOaGafqiVd0MbaKaadaWgaaWcbaGaamOA aaqabaGccaGGSaGafyyeIuUbaKaadaWgaaWcbaGaamOAaaqabaGcca GGPaGaaiilaiaadQgacqGH9aqpcaaIXaGaaiilaiaaikdaaaa@4589@ , where μ ^ = ( μ ^ j1 , μ ^ j2 ,..., μ ^ jT ) T MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqbeY7aTzaaja Gaeyypa0JaaiikaiqbeY7aTzaajaWaaSbaaSqaaiaadQgacaaIXaaa beaakiaacYcacuaH8oqBgaqcamaaBaaaleaacaWGQbGaaGOmaaqaba GccaGGSaGaaiOlaiaac6cacaGGUaGaaiilaiqbeY7aTzaajaWaaSba aSqaaiaadQgacaWGubaabeaakiaacMcadaahaaWcbeqaaiaadsfaaa aaaa@4B70@  and ^ j MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqbggHiLBaaja WaaSbaaSqaaiaadQgaaeqaaaaa@39DD@  of T×T MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadsfacqGHxd aTcaWGubaaaa@3AD7@  covariance matrix are moment estimates of μ j MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiabeY7aTnaaBa aaleaacaWGQbaabeaaaaa@39DF@  and j MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiabggHiLpaaBa aaleaacaWGQbaabeaaaaa@39CD@  and for each bootstrap sample f ^ 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadAgagaqcam aaBaaaleaacaaIYaaabeaaaaa@38F1@ , the estimate of f 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadAgadaWgaa WcbaGaaGOmaaqabaaaaa@38E1@  is calculated.

The histograms and Q-Q plots are constructed using the bootstrap values of f ^ 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadAgagaqcam aaBaaaleaacaaIYaaabeaaaaa@38F1@ generated by both methods and are shown in Figure 3 & Figure 4 respectively.

Figure 3 Left panel: Histogram of 3000 parametric bootstrap of f ^ 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadAgagaqcam aaBaaaleaacaaIYaaabeaaaaa@38F1@ . Right panel: Histogram of 3000 non-parametric bootstrap of f ^ 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadAgagaqcam aaBaaaleaacaaIYaaabeaaaaa@38F1@ .
Figure 4 Left panel: Q-Q plot for the values of f ^ 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadAgagaqcam aaBaaaleaacaaIYaaabeaaaaa@38F1@ generated by parametric bootstrap method. Right panel: Q-Q plot for the values of f ^ 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadAgagaqcam aaBaaaleaacaaIYaaabeaaaaa@38F1@ generated by non-parametric bootstrap method.

The Bootstrap parametric and non-parametric sampling distributions of f ^ 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadAgagaqcam aaBaaaleaacaaIYaaabeaaaaa@38F1@ shown in the left and right panels of Figure 2 are almost symmetrical. However, the non-parametric sampling distribution of the similarity factor f ^ 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadAgagaqcam aaBaaaleaacaaIYaaabeaaaaa@38F1@ is more symmetrical than that of the parametric one. In addition, the bootstrap parametric sampling distribution of f ^ 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadAgagaqcam aaBaaaleaacaaIYaaabeaaaaa@38F1@  is wider than the non-parametric bootstrapping. The percentile confidence methods work well if the underlying probability distribution from which samples are drawn is symmetric and the distribution of statistics is also symmetric. The reliability of the confidence interval for true parameter f 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadAgadaWgaa WcbaGaaGOmaaqabaaaaa@38E1@ by bootstrap method relies upon the symmetrical pattern of the sampling distribution of f ^ 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadAgagaqcam aaBaaaleaacaaIYaaabeaaaaa@38F1@ . The Q-Q plot of the sampling distribution of f ^ 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadAgagaqcam aaBaaaleaacaaIYaaabeaaaaa@38F1@ generated by a parametric bootstrap method, given in the left panel of Figure 3 confirms normality better than the Q-Q plot in the right panel generated by non-parametric bootstrap method. However, Q-Q plots do not confirm the normality assumption. So we apply more rigorous statistical test to verify the normality of the data. A commonly used normality test is Jargue-Bera test,3 which is based on skewness and kurtosis. In what follows we present some characteristics of the sampling distribution of f ^ 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadAgagaqcam aaBaaaleaacaaIYaaabeaaaaa@38F1@ obtained by both methods and the results of Jargue-Bera test. To assess basic properties of f ^ 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadAgagaqcam aaBaaaleaacaaIYaaabeaaaaa@38F1@ , we carry out empirical simulation study by Monte Carlo method. We estimate the characteristics of the sampling distribution defined in Section 3.2 with Monte Carlo size B=3000 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadkeacqGH9a qpcaaIZaGaaGimaiaaicdacaaIWaaaaa@3BC6@ . The simulation average (ME) of the statistics, and the coefficient of variation (CV), the ratio of the standard deviation of the statistic over the absolute value of ME based on 100 simulation replications, are presented in Table 2.

Method

ME

CV

Non-parametric

Mean

63.2

0.0009

Variance

4.24

0.0427

Coefficient of Skewness

0.01

0.4776

Coefficient of Kurtosis

3.25

0.0619

Bias

-0.43

0.1778

Parametric

Mean

63.15

0.0012

Variance

4.46

0.0476

Coefficient of Skewness

0.14

0.5938

Coefficient of Kurtosis

3.11

0.0619

Bias

-0.43

0.1776

Table 2 Summary measures of sampling distribution of under parametric and non-parametric bootstrap method

Both bootstrap estimators of the expected value of the sampling distribution f ^ 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadAgagaqcam aaBaaaleaacaaIYaaabeaaaaa@38F1@  are downward-biased. Non-parametric estimators are better than those obtained in parametric method in terms of variance and skewness. The coefficients of skewness in both procedures indicate that the sampling distribution is almost symmetric but slightly positively skewed. The coefficients of kurtosis under both procedures are slightly more than 3. The sampling distribution of f ^ 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadAgagaqcam aaBaaaleaacaaIYaaabeaaaaa@38F1@  is approximately normal. We calculate Jargue-Bera test statistic to check the normality of the distribution of f ^ MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadAgagaqcaa aa@3809@ . For a symmetric distribution, the third moment about mean ( μ 3 ) MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaamaabmaabaGaeq iVd02aaSbaaSqaaiaaiodaaeqaaaGccaGLOaGaayzkaaaaaa@3B40@ and coefficient of skewness ( τ ) MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaamaabmaabaGaeq iXdqhacaGLOaGaayzkaaaaaa@3A5C@ are equal to zero. The normal distribution is characterized with τ=0 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiabes8a0jabg2 da9iaaicdaaaa@3A93@ and coefficient of kurtosis, k=3 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadUgacqGH9a qpcaaIZaaaaa@39C1@ . A joint test of τ=0 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiabes8a0jabg2 da9iaaicdaaaa@3A93@  and k=3 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadUgacqGH9a qpcaaIZaaaaa@39C1@  is often used as a test of normality. Jargue & Bera26 proposed a statistic to test the normality of a distribution. Their proposed test statistic under the normality assumption is

JB=B( τ 2 6 + ( k3 ) 2 24 ) χ ˜ 2 2 , MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadQeacaWGcb Gaeyypa0JaamOqamaabmaabaWaaSaaaeaacqaHepaDdaahaaWcbeqa aiaaikdaaaaakeaacaaI2aaaaiabgUcaRmaalaaabaWaaeWaaeaaca WGRbGaeyOeI0IaaG4maaGaayjkaiaawMcaamaaCaaaleqabaGaaGOm aaaaaOqaaiaaikdacaaI0aaaaaGaayjkaiaawMcaaiqbeE8aJzaaia WaaSbaaSqaaiaaikdaaeqaaOWaaWbaaSqabeaacaaIYaaaaOGaaiil aaaa@4B5F@

where B is the number of the bootstrap samples. We can use this statistic to test the normality of the distribution of f ^ 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadAgagaqcam aaBaaaleaacaaIYaaabeaaaaa@38F1@ . We have JB=7.86 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadQeacaWGcb Gaeyypa0JaaG4naiaac6cacaaI4aGaaGOnaaaa@3C9F@  and 11.31 for parametric and non-parametric sampling distribution of f ^ 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadAgagaqcam aaBaaaleaacaaIYaaabeaaaaa@38F1@  respectively. These are highly insignificant ( χ 2 2 ( 0.05 )=5.991 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiabeE8aJnaaDa aaleaacaaIYaaabaGaaGOmaaaakmaabmaabaGaaGimaiaac6cacaaI WaGaaGynaaGaayjkaiaawMcaaiabg2da9iaaiwdacaGGUaGaaGyoai aaiMdacaaIXaaaaa@439A@  ) and ( χ 2 2 ( 0.01 )=9.21 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiabeE8aJnaaDa aaleaacaaIYaaabaGaaGOmaaaakmaabmaabaGaaGimaiaac6cacaaI WaGaaGymaaGaayjkaiaawMcaaiabg2da9iaaiMdacaGGUaGaaGOmai aaigdaaaa@42D0@ ). Thus we may conclude that the distribution of f ^ 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadAgagaqcam aaBaaaleaacaaIYaaabeaaaaa@38F1@ is not normal under both procedures, and we apply bootstrap algorithms to construct CIs for f 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaabaaaaaaaaape GaamOza8aadaWgaaWcbaWdbiaaikdaa8aabeaaaaa@392F@ .

Confidence Intervals for f 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaabaaaaaaaaape GaamOza8aadaWgaaWcbaWdbiaaikdaa8aabeaaaaa@392F@

For each bootstrap method of sampling, bootstrap-t, percentile, bias-corrected, and bias-corrected and accelerated confidence intervals for the parameter f 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaabaaaaaaaaape GaamOza8aadaWgaaWcbaWdbiaaikdaa8aabeaaaaa@392F@ are constructed and presented in Table 3.

Method

500 Bootstraps

1000 Bootstraps

1500 Bootstraps

2000 Bootstraps

2500 Bootstraps

Non-parametric

Mean

CI

Mean

CI

Mean

CI

Mean

CI

Mean

CI

Bootstrap-t

63.17

(60.75-67.59)

63.21

(60.66-67.40)

63.21

(60.67-67.29)

63.19

(60.63-67-04)

63.21

(60.73-67.24)

BP

(59.88,66.48)

(59.83,66.37)

(59.89,66.44)

(59.75,66.55)

(59.74,66.48)

BC

(60.72-67.49)

(60.91-67.49)

(60.66-67.18)

(60.75-67.24)

(60.78-67.30)

BCa

(60.86,66.99)

(60.58,67.22)

(60.70,67.36)

60.64,67.29)

(60.65,67.31)

Parametric

Bootstrap-t

63.12

(60.88-67.30)

63.09

(60.76-67.49)

63.13

(60.68-67.64)

63.19

(60.54-67.58)

63.15

(60.73-67.63)

BP

(59.51-66.49)

(59.74-66.80)

(59.54-66.47)

(59.71-66.48)

(59.64-66.49)

BC

(60.52-67.22)

(60.41-67.29)

(60.42-67.01)

(60.45-67.14)

(60.33-67.14)

BCa

(60.61-67.27)

(60.46-67.32)

(60.44-67.04)

(60.49-67.23)

(60.36-67.19)

Table 3 Non-parametric and Parametric Bootstrap Confidence Intervals for

The observed value of f ^ 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadAgagaqcam aaBaaaleaacaaIYaaabeaaaaa@38F1@  for the original dissolution data is 63.58. At 10% average distance at all time-points the similarity criterion is 50. Since the point estimate of f 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadAgadaWgaa WcbaGaaGOmaaqabaaaaa@38E1@ , f ^ 2 =63.58 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadAgagaqcam aaBaaaleaacaaIYaaabeaakiabg2da9iaaiAdacaaIZaGaaiOlaiaa iwdacaaI4aaaaa@3DB1@  is greater than the criterion value of 50, two drugs are considered to be same in terms of average dissolution data. The table 3 shows the 90% confidence interval for E( f ^ 2 ) MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadweadaqada qaaiqadAgagaqcamaaBaaaleaacaaIYaaabeaaaOGaayjkaiaawMca aaaa@3B4E@  with the bootstrap replications 500:500:2500 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqiVu0Je9sqqrpepC0xbbL8F4rqqrFfpeea0xe9Lq=Jc9 vqaqpepm0xbba9pwe9Q8fs0=yqaqpepae9pg0FirpepeKkFr0xfr=x fr=xb9adbaqaaeGaciGaaiaabeqaamaabaabaaGcbaqcLbsacaaI1a GaaGimaiaaicdacaGG6aGaaGynaiaaicdacaaIWaGaaiOoaiaaikda caaI1aGaaGimaiaaicdaaaa@3F56@ .

Under parametric and non-parametric bootstrap sampling schemes and all the bootstrap CI methods, the 90% lower confidence interval for E( f ^ 2 ) MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadweadaqada qaaiqadAgagaqcamaaBaaaleaacaaIYaaabeaaaOGaayjkaiaawMca aaaa@3B4E@  is greater than the similarity criterion value 50. This indicates that two drugs are similar.

However, under both parametric and non-parametric approaches, the percentile confidence interval is wider than the other bootstrap confidence intervals. This method, however, does not incorporate the skewness of the sampling distribution of f ^ 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadAgagaqcam aaBaaaleaacaaIYaaabeaaaaa@38F1@ . The BCa method corrects the bias and skewness in the sampling distribution of statistic. In the setting of both parametric and non-parametric procedures BCa gives the shortest confidence interval for E( f ^ 2 ) MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadweadaqada qaaiqadAgagaqcamaaBaaaleaacaaIYaaabeaaaOGaayjkaiaawMca aaaa@3B4E@ . The accuracy of the parametric and non-parametric bootstrap approximate confidence intervals for E( f ^ 2 ) MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadweadaqada qaaiqadAgagaqcamaaBaaaleaacaaIYaaabeaaaOGaayjkaiaawMca aaaa@3B4E@  cannot be accessed directly just by eyeballing. To see which method works well in our situation, we perform empirical comparisons of these bootstrap confidence intervals in the next section.

Empirical comparisons

In this section we examine and compare Bootstrap-t, BP, BC, and BCa confidence sets using simulation approach. The Monte Carlo simulation with size 500 is used to calculate the simulation average of a confidence bound (AV), and the simulation estimates of the expected length of a two-sided confidence interval (EL). This simulation study is performed under both non-parametric and parametric bootstrap sampling schemes.

Table 4 shows average left and right endpoints of the confidence intervals constructed by various methods and the expected length of two sided confidence intervals for E( f ^ 2 ) MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadweadaqada qaaiqadAgagaqcamaaBaaaleaacaaIYaaabeaaaOGaayjkaiaawMca aaaa@3B4E@ . All the methods under the non-parametric bootstrap sampling scheme provide shorter confidence intervals than those obtained under the parametric scheme. The lower confidence bounds for percentile methods under both bootstrapping schemes shift more to the left compared to those obtained by other methods. The bootstrap BCa confidence interval has the smallest expected length, capturing the asymmetry of the exact confidence intervals (CI).28

Method

Left

Right

Two-sided

AV

AV

EL

Non-Parametric

Bootstrap-t

60.61

67.36

6.75

BP

59.81

66.56

6.75

BC

60.65

67.39

6.74

BCa

60.7

67.43

6.73

Parametric

Bootstrap-t

60.66

67.56

6.9

BP

59.61

66.51

6.9

BC

60.45

67.24

6.81

BCa

60.5

67.31

6.77

Table 4 Comparison of the bootstrap-t, bootstrap percentile (BP), Bias-corrected (BC) and BCa confidence sets for E( f ^ 2 ) MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadweadaqada qaaiqadAgagaqcamaaBaaaleaacaaIYaaabeaaaOGaayjkaiaawMca aaaa@3B4E@ ( 1α )=0.90 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqiVu0Je9sqqrpepC0xbbL8F4rqqrFfpeea0xe9Lq=Jc9 vqaqpepm0xbba9pwe9Q8fs0=yqaqpepae9pg0FirpepeKkFr0xfr=x fr=xb9adbaqaaeGaciGaaiaabeqaamaabaabaaGcbaqcfa4aaeWaaO qaaKqzGeGaaGymaiabgkHiTiabeg7aHbGccaGLOaGaayzkaaqcLbsa cqGH9aqpcaaIWaGaaiOlaiaaiMdacaaIWaaaaa@4075@

Conclusion

In this study parametric and non-parametric resampling methods are used to explore statistical properties of the similarity factor f 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadAgadaWgaa WcbaGaaGOmaaqabaaaaa@38E1@ . Under these two methods, 90% confidence intervals are constructed using different bootstrap approaches for the true expected value of f ^ 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiqadAgagaqcam aaBaaaleaacaaIYaaabeaaaaa@38F1@ .

For small sample sizes as in this study ( n 1 =12 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaad6gadaWgaa WcbaGaaGymaaqabaGccqGH9aqpcaaIXaGaaGOmaaaa@3B6F@ for test drug, and n 2 =12 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaad6gadaWgaa WcbaGaaGOmaaqabaGccqGH9aqpcaaIXaGaaGOmaaaa@3B70@ for reference drug), nonparametric bootstrapping provides relatively smaller expected length of confidence intervals for the parameter f 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadAgadaWgaa WcbaGaaGOmaaqabaaaaa@38E1@  compared to those obtained by the parametric method. However, the parametric bootstrap usually performs well over the non-parametric bootstrap method for small sample. In our study both methods provide similar CIs with no substantial advantage for considering one over the other. This may be due to the fact that the observed distribution of the reference drug was not normal. But we treated the observed distribution of the reference drug as normal to facilitate the parametric approach.

In addition to larger expected length confidence intervals, the parametric bootstrap method also provides less stable moment estimators of f 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadAgadaWgaa WcbaGaaGOmaaqabaaaaa@38E1@  compared to the non-parametric bootstrapping methods. We note that BCa performs the best in terms of producing smaller expected length among all the algorithms under both schemes. However, for small samples we recommend constructing bootstrap confidence intervals using non-parametric methods since these methods produce better results.

In this article, we showed that bootstrap is a powerful and effective means of setting approximate confidence intervals for the dissolution similarity measure f 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadAgadaWgaa WcbaGaaGOmaaqabaaaaa@38E1@  using several computing algorithms. These results have policy implications for regulatory agencies such as FDA. Confidence intervals for the similarity factor f 2 MathType@MTEF@5@5@+= feaagKart1ev2aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr 4rNCHbGeaGqkY=Mj0xXdbba91rFfpec8Eeeu0xXdbba9frFj0=OqFf ea0dXdd9vqaq=JfrVkFHe9pgea0dXdar=Jb9hs0dXdbPYxe9vr0=vr 0=vqpWqaaeaabiGaciaacaqabeaadaqaaqaaaOqaaiaadAgadaWgaa WcbaGaaGOmaaqabaaaaa@38E1@  provide more reliable prediction on the similarity of dissolution profiles of the test and the reference drugs.

Acknowledgements

None.

Conflict of interests

Author declares that there is no conflict of interest.

References

  1. Adams E. Evaluation of dissolution profiles using principal component analysis. International Journal of Pharmaceutics. 2001;212(1):41‒53.
  2. Huey Lin Ju. On the assessment of similarity of drug dissolution profiles‒A simulation study. Drug Information Journal. 1997;31(4):1273‒1289.
  3. Vinod PS. In vitro dissolution profile comparison:‒Statistics and analysis of the similarity factor, f2. Pharmaceutical Research. 1998;15(6):889‒896.
  4. Dawoodbhai S, Suryanarayan E, Woodruff C, et al. Optimization of tablet formulation containing talc. Drug Develop Industrial Pharmacy. 1991;17(10):1343‒1371.
  5. Pena Romero A, Caramella C, Ronchi M. Water uptake and force development in an optimized prolonged release formulation. International Journal Pharmaceutics. 1991;73:239‒248.
  6. Langenbucher F. Linearization of dissolution rate curves by Weilbull distribution. 1972.
  7. Kervinen L, Yliruusi J. Modeling S‒shaped dissolution curves. International Journal of Pharmaceuticals. 1993;92(1‒3):115‒122.
  8. Tsong Y, Hammerstrom T, Lin KK, et al. The dissolution testing sampling acceptance rules. Journal of Biopharmaceutical Statistics. 1995;5(2):171‒184.
  9. Tsong Y, Hammerstrom T, Chen JJ. Multiple dissolution specification and acceptance sampling rule based on profile modeling and principal component analysis. J Biopharm Stat. 1997;7(3):423‒439.
  10. Moore JR, Flanner HH. Mathematical comparison of curves with an emphasis on dissolution profiles. Pharma Technol. 1996;20(6):64‒74.
  11. Chow SC, Liu JP. Design and analysis of bioavailability and bioequivalence studies. Chapman and Hall/CRC, 2008.
  12. Tsong Y. Statistical assessment of mean differences between two dissolution datasets. Rockville: Presented at the 1995 Drug Information Association Workshop; 1995.
  13. Tsong Y, Hammerstrom T, Sathe P, et al. Statistical assessment of mean differences between two dissolution data sets. Drug Information Journal. 1996;30(4):1105‒1112.
  14. DiCiccio TJ, Romano JP. A review of bootstrap confidence Intervals. Journal of Royal Statistical Society. Series B. 1988;50(3): 338‒354.
  15. Efron B. Bootstrap methods: Another look at the Jackknife. Annals of Statistics. 1979;7(1):1‒26.
  16. Chernick MR. Bootstrap Methods: A Guide for Practitioners and Researchers. 2nd edn. Wiley: Hoboken; 2008.
  17. Davison AC, Hinkley DV. Bootstrap Methods and Their Applications. Cambridge University press, Cambridge: New York; 1997.
  18. DiCiccio TJ, Efron B. Bootstrap confidence intervals. Statistical Science. 1996;11(3):189‒212.
  19. Efron B, Tibshirani R. An Introduction to the Bootstrap. , New York: Chapman and Hall; 1994.
  20. Efron B. The Jackknife, the Bootstrap, and Other Resampling Plans. Philadelphia, Society for Industrial and Applied Mathematics, 1982.
  21. Schenker N. Qualms about bootstrap confidence intervals. Journal of American Statistical Association. 1985;80(390):360‒361.
  22. Efron, B. (1981): Nonparametric standard errors and confidence intervals. Canadian Journal of Statistics, 9, 139‒172.
  23. Efron B. Better bootstrap confidence intervals. Journal of the American Statistical Association. 1981;82(39):171‒185.
  24. Bickel Pj, Freedman DA. Some asymptotics on the bootstrap. Annals of Statistics. 1981;9(6):1196‒1217.
  25. Hall P. The Bootstrap and the Edgeworth Expansion. New York: Springer‒Verlag Hall; 1992.
  26. Efron B. Nonparametric standard errors and confidence intervals. Canadian Journal of Statistics. 1981;9(2):139‒172.
  27. Carlos M Jarque, Anil K Bera. A test for normality of observations and regression residuals. International Statistical Review. 1987;55(2):163‒172.
  28. Tsong Y, Hammerstrom T. Statistical issues in drug quality control based on dissolution testing. Proceedings of the Biopharmaceutical Section of the American Statistical Association. 1992; 295‒300.
Creative Commons Attribution License

©2018 Islam, et al. This is an open access article distributed under the terms of the, which permits unrestricted use, distribution, and build upon your work non-commercially.