Testing equality of two-parameter exponentially distributed mean lifetimes under unequal failure rates

Abstract

Testing the equality of means of several skewed populations, particularly in the presence of nuisance parameters, is a central challenge in statistics. While various tests have been proposed for such as log-normal, inverse-normal, and exponential distributions leveraging methods like generalized p-value, parametric bootstrap, and the fiducial approach, there remains a notable gap in the literature, the absence of a computational approach method-based test for the two-parameter exponential distribution. Such a method is essential for achieving robust results in small sample sizes while considering power and Type I error probability. In response to this gap, our paper introduces and implements novel compu-tational approach tests embedded in the doex package in R. Our focus is on assessing the equality of means for several skewed populations following a two-parameter exponential distribution. We conduct a comprehensive comparison of our proposed tests against existing alternatives, evaluating their penalized power and Type I error probability. Notably, our computational approach tests exhibit superior performance, particularly in cases involving small samples and balanced designs. Furthermore, to illustrate the practical relevance of our proposed tests, we present a real-world application using authentic data. This empirical demonstration serves to underscore the efficacy and applicability of our novel computational approach tests in real-world scenarios.

Keywords

• Computational Approach
• Non-Normal ANOVA
• Penalized Power
• Unequal Scales

References

1. REFERENCES
2. [1] Weerahandi S. ANOVA under unequal error variances. Biometrics 1995;51:589–599. [CrossRef]
3. [2] Krishnamoorthy K, Lu F, Mathew T. A parametric bootstrap approach for ANOVA with unequal variances: Fixed and random models. Comput Stat Data Anal 2007;51:5731–5742. [CrossRef]
4. [3] Li X, Wang J, Liang H. Comparison of several means: A fiducial based approach. Comput Stat Data Anal 2011;55:1993–2002. [CrossRef]
5. [4] Gokpinar EY, Gokpinar F. A test based on the computational approach for equality of means under the unequal variance assumption. Hacettepe J Math Stat 2012;41:605–613.
6. [5] Gokpinar EY, Gokpinar F. Testing the equality of several log-normal means based on a computational approach. Commun Stat Simul Comput 2017;46:1998–2010. [CrossRef]
7. [6] Jafari AA, Abdollahnezhad K. Testing the equality means of several lognormal distributions, Commun Stat Simul Comput 2017;46:2311–2320. [CrossRef]
8. [7] Weerahandi S, Krishnamoorthy K. A note reconciling ANOVA tests under unequal error variances. Comput Stat Data Anal 2019;55:1993–2002. [CrossRef]

9. [8] Chen HJ. A new range statistic for comparisons of several exponential location parameters. Biometrika 1982;69:257–260. [CrossRef]
10. [9] Singh N. The likelihood ratio test for the equality of location parameters of exponential populations based on Type II censored samples. Technometrics 1983;25:193–195. [CrossRef]
11. [10] Kambo NS, Awad AM. Testing equality of location parameters of k exponential distributions. Commun Stat Theory Methods 1985;14:567–583.
12. [11] Hsieh HK. An exact test for comparing location parameters of k exponential distributions with unequal scaled based on Type II censored data. Technometrics 1986;28:157–164. [CrossRef]
13. [12] Vaughan DC, Tiku ML. Testing the equality of location parameters of exponential populations from censored samples. Commun Stat Theory Methods 1993;22:2567–2581. [CrossRef]
14. [13] Tiku ML, Vaughan DC. Testing equality of location parameters of two exponential distributions from censored samples. Commun Stat Theory Methods 1991;20:929–944. [CrossRef]
15. [14] Ananda MMA, Weerahandi S. Testing the difference of two exponential means using generalized p-values. Commun Stat Simul Comput 1996;25:521–532. [CrossRef]
16. [15] Wu SF. One stage multiple comparisons with the control for exponential mean lifetimes based on doubly censored samples under heteroscedasticity. Commun Stat Simul Comput 2021;50:1473–1483. [CrossRef]
17. [16] Malekzadeh A, Jafari AA. Inference on the equality means of several two-parameter exponential distributions under progressively Type II censoring, Commun Stat Simul Comput 2020;49:3196–3211. [CrossRef]
18. [17] Johnson NL, Kotz S. Continuous Univarite Distributions. Boston: Houghton Miffin Company; 1970.
19. [18] Raiz M, Kumar A, Mishra VN, Rao N. Dunkl analogue of Schurer-Beta operators and their approximation behaviour. Math Found Comput 2022;5:315–330. [CrossRef]
20. [19] Rajawat RS, Singh KK, Mishra VN. Approximation by modified Bernstein polynomials based on real parameters. Math Found Comput 2024;7:297–309. [CrossRef]
21. [20] Cochran WG. Problem arising in the analysis of a series of similar experiments. J Royal Stat Soc 1937;4:102–118. [CrossRef]
22. [21] Tsui KW, Weerahandi S. Generalized p-value in significance testing of hypothesis in the presence of nuisance parameters. J Am Stat Assoc 1989;84:602–607. [CrossRef]
23. [22] Pal N, Lim WK, Ling CH. A computational approach to statistical inferences. J Appl Probab Stat 2007;2:13–35.
24. [23] Zheng M. Penalized maximum likelihood estimation of two-parameter exponential distributions. Minnesota, USA: The Faculty of Graduate School of the University of Minnesota; 2013.
25. [24] Cavus M, Yazici B, Sezer A. Penalized power approach to compare the power of the tests when Type I error probabilities are different. Commun Stat Simul Comput 2021;50:1912–1926. [CrossRef]
26. [25] Cavus M, Yazici B. Doex: One-way heteroscedastic ANOVA tests. R package: ver.1.3, 2020.
27. [26] Cavus M, Yazici B. Testing the equality of normal distributed and independent groups’ means under unequal variances by doex package. The R J 2021;12:134–154. [CrossRef]
28. [27] Box GEP, Cox DR. An analysis of transformation. J Royal Stat Soc 1964;26:211–252. [CrossRef]
29. [28] Cavus M, Yazici B, Sezer A. Modified tests for comparison of group means under heteroskedasticity and non-normality caused by outlier(s). Hacettepe J Math Stat 2017;46:493–510. [CrossRef]
30. [29] Cavus M, Yazici B, Sezer A. Analyzing regional export data by the modified generalized F-test. Int J Econ Administr Stud 2018:541–552.
31. [30] Cavus M, Yazici B, Sezer A. Penalized power approach to compare the power of the tests when Type I error probabilities are different. Commun Stat Simul Comput 2021;50:1912–1926. [CrossRef]