In the simulation study, various experimental designs are considered to evaluate the performance of proposed modified Welch classical ANOVA tests in terms of the type I-errors studies via simulation study.
📌 3 Key Takeaways
🎯
Objective & Hypothesis:A modification to Welch test statistic is proposed to test the equality of population means of various groups under a Weibull distribution.
📊
Key Findings & Data:The proposed test statistic is simple and corresponds to the standard Welch test statistic in which the maximum likelihood mean and variance estimators are replaced with robust estimators based on quantile, quantile least square and repeated median.
💡
Conclusion & Impact:In the simulation study, various experimental designs are considered to evaluate the performance of proposed modified Welch classical ANOVA tests in terms of the type I-errors studies via simulation study.
🔬Design:Research Article
📍Scope:Comprehensive Sample
⚙️Variables:ANOVA · Welch Test Statistic · Robustness of estimators
Abstract
A modification to Welch test statistic is proposed to test the equality of population means of various groups under a Weibull distribution. The proposed test statistic is simple and corresponds to the standard Welch test statistic in which the maximum likelihood mean and variance estimators are replaced with robust estimators based on quantile, quantile least square and repeated median. The influence function and breakdown point of these robust estimators are obtained to show their robustness properties. In the simulation study, various experimental designs are considered to evaluate the performance of proposed modified Welch classical ANOVA tests in terms of the type I-errors studies via simulation study.
Karagöz, D. (2016). Modified Welch test statistic for ANOVA under Weibull distribution. Hacettepe Journal of Mathematics and Statistics, Volume: 45 Issue: 2. https://ajindex.com/articles/1403179
Abernethy, R.B., Breneman, J.E., Medlin, C.H. and Reinman, G.L., 1983, Weibull Analysis Handbook, Technical Report, AFWAL-TR-83- 207, Air Force Wright Aeronautical Laboratories, Washington, D.C., http://handle.dtic.mil/100.2/ADA143100.
Boudt, K., Caliskan, D. and Croux, C., 2011, Robust explicit estimators of Weibull parameters, Metrika, 73, 187-209.
Cochran, W.G., 1937, Problems arising in the analysis of a series of similar experimentals, Journal of Royal Statistics Soc.Supp., 4, 102-118.
Cohen, A. C. and Whitten, B. J., 1988, Parameter Estimation in Reliability and Life Span Models, New York: Marcel Dekker.
DerSimonian, R. and Laird, N., 1986, Meta-analysis in clinical trials, Controlled Clin. Trials, 7, 177-188.
Gupta, R.D. and Kundu, D., 2001, Exponentiated Exponential Family: An Alternative to Gamma and Weibull Distributions, Biometrical Journal 43,1, 117-130
James, G., S., 1951, The comparison of several groups of observations when the ratios of the population variances are unkown, Biometrika, 38, 324-329.
Karagöz, D., 2012, Robust Analaysis of Variance, PhD thesis: Hacettepe University, Ankara, Turkey.
Kulinskaya, E., Staudte, R.G. and Gao, H., 2003, Power approximations in testing for unequal means in a One-Way ANOVA weighted for unequal variances, Communications in Statistics Theory and Methods, 32,2353-2371.
Kulinskaya, E. and Dollinger, M.B., 2007, Robust weighted One-Way ANOVA: Improved approximation and efficiency, Journal of Statistical Planning and Inference, 137, 462-472.
Şenoğlu, B., 2005, Robust 2 k Factorial Design with Weibull Error Distributions, Journal of Applied Statistics, 32-10, 1051-1066.
Takkouche, B., Cadarso-Suarez, C. and Spiegelman, D., 1999, Evaluation of old and new tests of heterogeneity in epidemiologic meta-analysis, Amer.J. Epidemiology 150, 206-215.
Welch, B.L., 1951, On the comparison of several mean values: an alternative approach. Biometrika, 38, 330-336.
Wilcox, R.R., 1995, The practical importance of heteroscedastic methods, using trimmed means versus means, and designing simulation studies, British J. Math.Statist.Psych., 48, 99-114.
Wilcox, R.R., 1997, Introduction to Robust Estimation and Hypothesis Testing, Academic Press, New York.