Testing the equality of two normal percentiles

Two statistics are suggested for testing the equality of two normal percentiles where population means and variances are unknown. The first is based on the generalized likelihood ratio test (LRT), the second on Cochran's statistic used in the Behrens-Fisher problem. Size and power comparisons are made by using simulation and asympototic theory.     

One-way ANOVA with Unequal Variances

We consider the one-way ANOVA problem of testing the equality of several normal means when the variances are not assumed to be equal. This is a generalization of the Behrens-Fisher problem, but even in this special case there is no exact test and the actual size of any test depends on the values of the nuisance parameters. Therefore, controlling the actual size of the test is of main concern. In this article, we first consider a test using the concept of generalized p-value. Extensive simulation studies show that the actual size of this test does not exceed the nominal level, for practically all values of the nuisance parameters, but the test is not too conservative either, in the sense that the actual size of the test can be very close to the nominal level for some values of the nuisance parameters. We then use this test to propose a simple F-test, which has similar properties but avoids the computations associated with generalized p-values. Because of its simplicity, both conceptually as well as computationally, this F-test may be more useful in practice, since one-way ANOVA is widely used by practitioners who may not be familiar with the generalized p-value and its computational aspects.     

An adjusted likelihood-ratio algorithm for the Behrens-Fisher problem

We present an algorithm which will incorporate an adjusted likelihood and a scaled likelihood procedure for the Behrens-Fisher problem. Nonuniqueness of the maximum likelihood estimates for the common mean are demonstrated on a simple example showing how ill-conditioned the maximum likelihood equations can be. Our results, however, show that the significance levels are stable even when the means are sensitive to perturbations.     

Exact tests based on the Baumgartner-Weiß-Schindler statistic—A survey

It is the purpose of this paper to review recently-proposed exact tests based on the Baumgartner-Weiß-Schindler statistic and its modification. Except for the generalized Behrens-Fisher problem, these tests are broadly applicable, and they can be used to compare two groups irrespective of whether or not ties occur. In addition, a nonparametric trend test and a trend test for binomial proportions are possible. These exact tests are preferable to commonly-applied tests, such as the Wilcoxon rank sum test, in terms of both type I error rate and power.     

The generalized inference on the sign testing problem about the normal variances

For the sign testing problem about the normal variances, we develop the heuristic testing procedure based on the concept of generalized test variable and generalized p-value. A detailed simulation study is conducted to empirically investigate the performance of the proposed method. Through the simulation study, especially in small sample sizes, the proposed test not only adequately controls empirical size at the nominal level, but also uniformly more powerful than likelihood ratio test, Gutmann's test, Li and Sinha's test and Liu and Chan's test, showing that the proposed method can be recommended in practice. The proposed method is illustrated with the published data.     

A Correction Note on Improvement in Variance Estimation Using Auxiliary Information

ABSTRACT

In this paper, the testing problem for homogeneity in the mixture exponential family is considered. The model is irregular in the sense that each interest parameter forms a part of the null hypothesis (sub-null hypothesis) and the null hypothesis is the union of the sub-null hypotheses. The generalized likelihood ratio test does not distinguish between the sub-null hypotheses. The Supplementary Score Test is proposed by combining two orthogonalized score tests obtained corresponding to the two sub-null hypotheses after proper reparameterization. The test is easy to design and performs better than the generalized likelihood ratio test and other alternative tests by numerical comparisons.     

Coefficients of lee-gurland two-sample test on normal means

The Behrens-Fisher problem in comparing means of two normal populations is revisited Lee and Gurland (1975) suggested a solution to the problem and provided the set of coefficients required in computing critical values for the case α=005, where α is the nominal level of significance This solution, called the Lee-Guiland Test in this article, has proven to be practical as far as calculation is involved, and more importantly, it maintains the actual size very close to α= 0.05 for possible values of the ratio of population variances This merit has not been attained by most of the Behrens-Fisher solutions in the literature. In this article, the coefficients for other values of α, namely 0 025, 0 01 and 0.005 are provided for wider applications of the test Moreover, careful and detailed comparisons are made in terms of size and power with the other practical solution:the Welch's Approximate t I est Due to a possible drawback of the Welch's Approximate t I est in controlling the actual size, especially for small a and small sample sizes, the Lee-Gurland lest presents itself as a slightly better' alternative in testing equality of two normal population means I he coefficients mentioned above are also fitted by the functions of the reciprocals of the degrees of freedom, so that the substantial amount of table-looking can be avoided Some discussions are also made in regarding the recent “Welch vs Gosset” argument: Should the Student's t Test be dispensed off’from the routine use in testing the equality of two normal means?.     

Testing equality of generalized variances of k multivariate normal populations

Generalized variance is a measure of dispersion of multivariate data. Comparison of dispersion of multivariate data is one of the favorite issues for multivariate quality control, generalized homogeneity of multidimensional scatter, etc. In this article, the problem of testing equality of generalized variances of k multivariate normal populations by using the Bartlett's modified likelihood ratio test (BMLRT) is proposed. Simulations to compare the Type I error rate and power of the BMLRT and the likelihood ratio test (LRT) methods are performed. These simulations show that the BMLRT method has a better chi-square approximation under the null hypothesis. Finally, a practical example is given.     

An adjusted likelihood-ratio approach to the behrens-fisher problem

In many situations it is necessary to test the equality of the means of two normal populations when the variances are unknown and unequal. This paper studies the celebrated and controversial Behrens-Fisher problem via an adjusted likelihood-ratio test using the maximum likelihood estimates of the parameters under both the null and the alternative models. This procedure allows the significance level to be adjusted in accordance with the degrees of freedom to balance the risk due to the bias in using the maximum likelihood estimates and the risk due to the increase of variance. A large scale Monte Carlo investigation is carried out to show that -2 InA has an empirical chi-square distribution with fractional degrees of freedom instead of a chi-square distribution with one degree of freedom. Also Monte Carlo power curves are investigated under several different conditions to evaluate the performances of several conventional procedures with that of this procedure with respect to control over Type I errors and power.     

Testing the equality of quantiles for several normal populations

Many procedures exist for testing equality of means or medians to compare several independent distributions. However, the mean or median do not determine the entire distribution. In this article, we propose a new small-sample modification of the likelihood ratio test for testing the equality of the quantiles of several normal distributions. The merits of the proposed test are numerically compared with the existing tests—a generalized p-value method and likelihood ratio test—with respect to their sizes and powers. The simulation results demonstrate that proposed method is satisfactory; its actual size is very close to the nominal level. We illustrate these approaches using two real examples.     

A fitted test for the behrens-fisher problem

In 1975, Lee and Gurland proposed a solution to the Behrens-Fisher problem. It had excellent control of size and power and was relatively simple to use. However it requires extensive special tables. This article proposes a modification of this approach. It replaces the tables with easily computed functions of the sample sizes and a standard t table. Control of size and power are equivalent to that obtained by Lee and Gurland. Furthermore, the test is also compared with the Welch's approximate t test and shows better control of size, with similar power curves when sample sizes are at least four from each of the two normal populations.     

A four-moment approach and other practical solutions to the behrens-f1sher problem

Fixed sample size approximately similar tests for the Behrens-Fisher problem are studied and compared with various other tests suggested in current sttistical methodelogy texts. Several fourmoment approxiamtely similar tests are developed and offered as alternatives. These tests are shown to be good practical solutions which are easily implemented in practice.     

Some tests with specified size for the behrens-fisher problem

For a given significance level α, Welch's approximate t-test for the Behrens-Fisher Problem is modified to get a test with size α. A useful result for carrying out the Berger and Boos test is provided. Simulation results give power comparisons of several size α tests.     

Test for Generalized Variance in Factor Analysis Model

We derive a likelihood ratio test for generalized variance a in factor analysis model. The asymptotic distribution of the test statistic follows chi-square distribution with one degree of freedom from a general theory of likelihood ratio test.     

Testing Homogeneity of Inverse Gaussian Scale Parameters Based on Generalized Likelihood Ratio

This article presents a new procedure for testing homogeneity of scale parameters from k independent inverse Gaussian populations. Based on the idea of generalized likelihood ratio method, a new generalized p-value is derived. Some simulation results are presented to compare the performance of the proposed method and existing methods. Numerical results show that the proposed test has good size and power performance.     

Mutual information and redundancy for categorical data

Most methods for describing the relationship among random variables require specific probability distributions and some assumptions concerning random variables. Mutual information, based on entropy to measure the dependency among random variables, does not need any specific distribution and assumptions. Redundancy, which is an analogous version of mutual information, is also proposed as a method. In this paper, the concepts of redundancy and mutual information are explored as applied to multi-dimensional categorical data. We found that mutual information and redundancy for categorical data can be expressed as a function of the generalized likelihood ratio statistic under several kinds of independent log-linear models. As a consequence, mutual information and redundancy can also be used to analyze contingency tables stochastically. Whereas the generalized likelihood ratio statistic to test the goodness-of-fit of the log-linear models is sensitive to the sample size, the redundancy for categorical data does not depend on sample size but depends on its cell probabilities.     

Hypothesis Testing of Hazard Ratio Parameters in Marginal Models for Multivariate Failure Time Data

Marginal hazard models for multivariate failure time data have been studied extensively in recent literature. However, standard hypothesis test statistics based on the likelihood method are not exactly appropriate for this kind of model. In this paper, extensions of the three commonly used likelihood hypothesis test statistics are discussed. Generalized Wald, generalized score and generalized likelihood ratio tests for hazard ratio parameters in a marginal hazard model for multivariate failure time data are proposed and their asymptotic distributions examined. The finite sample properties of these statistics are studied through simulations. The proposed method is applied to data from Busselton Population Health Surveys.     

A Test for the Equality of Parameters for Separate Regression Models in the Presence of Heteroskedasticity

Testing for the equality of regression coefficients across two regressions is a problem considered by analysts in a variety of fields. If the variances of the errors of the two regressions are not equal, then it is known that the standard large sample F-test used to test the equality of the coefficients is compromised by the fact that its actual size can differ substantially from the stated level of significance in small samples. This article addresses this problem and borrows from the literature on the Behrens-Fisher problem to provide some simple modifications of the large sample test which allows one to better control the probability of committing a Type I error. Empirical evidence is presented which indicates that the suggested modifications provide tests which are superior to well-known alternative tests over a wide range of the parameter space.     

A sequential two sample t test using welch type modification for unequal variances

The sequential analogue of the Behrens-Fisher problem is considered, The pooled-variance two sample sequential t test is modified to account for unequal variances. Operating cha-racteristic and average-sample-number curves are calculated for both the pooled variance and the modified t tests by computer simulations, An example is given using data from a tissue assay for breast cancer tumors.     

Accurate tests for the equality of coefficients of variation

ABSTRACT

A frequently encountered statistical problem is to determine if the variability among k populations is heterogeneous. If the populations are measured using different scales, comparing variances may not be appropriate. In this case, comparing coefficient of variation (CV) can be used because CV is unitless. In this paper, a non-parametric test is introduced to test whether the CVs from k populations are different. With the assumption that the populations are independent normally distributed, the Miller test, Feltz and Miller test, saddlepoint-based test, log likelihood ratio test and the proposed simulated Bartlett-corrected log likelihood ratio test are derived. Simulation results show the extreme accuracy of the simulated Bartlett-corrected log likelihood ratio test if the model is correctly specified. If the model is mis-specified and the sample size is small, the proposed test still gives good results. However, with a mis-specified model and large sample size, the non-parametric test is recommended.