A note on the multiple comparisons of exponential location parameters with several controls under heteroscedasticity

Abstract

Several researchers have addressed the problem of constructing simultaneous confidence intervals (SCIs) for comparing exponential location parameters with a control or controls under heteroscedasticity when sample sizes are equal or unequal. They usually used simulation-based inference procedures or Lam's technique that leads to conservative SCIs. In this paper, we present a set of SCIs for comparing exponential location parameters with a control, controls and the best control under heteroscedasticity when sample sizes are possibly unequal. Our method is not a simulation-based inference procedure and our results show that the proposed SCIs have some advantages over others.

Keywords

• Average volume,Coverage probability,Conservative,Simulation

References

1. Chang, Y. P., Huang, W. T. Simultaneous ducial generalized condence intervals for all pairwise comparisons of means, International Journal of information and Management Sci- ences 20, 459-467, 2009.
2. Hannig, J., Iyer, H., Patterson, P. Fiducial generalized condence intervals, Journal of American Statistical Association 101, 254-269, 2006a.
3. Hannig, J., Lidong, E., Abdel-Karim, A., Iyer, H. Simultaneous ducial generalized con- dence intervals for ratios of means of Lognormal distributions, Austrian Journal of Statis- tics, 35, 261-269, 2006b.
4. Kharrati-Kopaei, M. A note on the simultaneous condence intervals for the successive dierences of exponential location parameters under heteroscedasticity, Statistical Method- ology 22, 1-7, 2015.
5. Kharrati-Kopaei, M., Malekzadeh, A., Sadooghi-Alvandi, S. M. Simultaneous ducial gener- alized condence intervals for the successive dierences of exponential location parameters under heteroscedasticity, Statistics and Probability Letters 83, 1547-1552, 2013.
6. Lam, K. Subset selection of normal population under heteroscedasticity, in: Proceedings of the second international advanced seminar/workshop on inference procedures associated with statistical ranking and selection, Sydney, Australia, 1987.
7. Lam, K. An improved two-stage selection procedure, Communications in Statistics, Simu- lation and Computation 17 (3), 995-1005, 1988.
8. Lam, K., Ng, C. K. Two-stage procedures for comparing several exponential populations with a control when scale parameters are unknown and unequal, Sequential Analysis 9 (2), 151-164, 1990.

9. Lawless, J. F., Singhal, K. Analysis of data from life test experiments under an exponential model, Naval Research Logistic Quarterly 27, 323-334, 1980.
10. Malekzadeh, A., Kharrati-Kopaei, M., Sadooghi-Alvandi, S. M. Comparing Exponential Location Parameters with Several Controls Under Heteroscedasticity, Computational Sta- tistics, DOI 10.1007/s00180-014-0481-6, 2014.
11. Maurya, V., Gill, A. N., Singh, P. Multiple comparisons with a control for exponential location parameters under heteroscedasticity, Journal of Applied Statistics 40, 1817-1830, 2013.
12. Maurya, V., Goyal, A., Gill, A. N. Multiple comparisons with more than one control for exponential location parameters under heteroscedasticity, Communications in Statistics- Simulation and Computation 40, 621-644, 2011.
13. Ng, C. K., Lam, K., Chen, H. J. Multiple comparisons of exponential location parameters with the best under type II censoring, American Journal of Mathematical and Management Sciences 12, 383-402, 1993.
14. Singh, P., Abebe, A. Comparing several exponential populations with more than one control, Statistical Methods and Applications 18, 359-374, 2009.
15. Wu, S., Lin, Y. P., Yu, Y. R. One-stage multiple comparisons with the control for exponential location parameters under heteroscedasticity, Computational Statistics and Data Analysis 54, 1372-1380, 2010.
16. Wu, S. F., Wu C. C. Two-stage multiple comparisons with the average for exponential location parameters under heteroscedasticity, Journal of Statistics and Planning Inference 134, 392-408, 2005.
17. Zelen, M. Application of exponential models problems in cancer research, Journal of Royal Statistical Society Series A (129), 368-398, 1966.