Advances and Applications in Statistics
Volume 53, Issue 3, Pages 259 - 283
(September 2018) http://dx.doi.org/10.17654/AS053030259 |
|
CONFIDENCE INTERVALS FOR VARIANCE AND DIFFERENCE BETWEEN VARIANCES OF ONE-PARAMETER EXPONENTIAL DISTRIBUTIONS
Warisa Thangjai, Sa-Aat Niwitpong and Suparat Niwitpong
|
Abstract: This paper proposes confidence intervals for the single variance and the difference between two exponential population variances. For single variance, generalized confidence interval, large sample confidence interval, and exact confidence interval were presented. The generalized confidence interval approach, large sample approach, and method of variance estimates recovery (MOVER) approach were used to construct confidence interval for difference between two variances. The coverage probability and average length of proposed confidence intervals were evaluated via the Monte Carlo simulation. The results indicated that the exact confidence interval provided much better performance than the other confidence intervals for the single variance, whereas the generalized confidence interval provided much better performance than the other confidence intervals for the difference between two variances. However, the MOVER confidence interval can be considered as an alternative to estimate the confidence interval for the difference between variances. Finally, two real data sets are given to illustrate proposed confidence intervals. |
Keywords and phrases: variance, exponential distribution, generalized confidence interval, large sample confidence interval, MOVER confidence interval. |
|
Number of Downloads: 408 | Number of Views: 3603 |
|