# zbMATH — the first resource for mathematics

Robust confidence interval for the variance. (English) Zbl 0918.62023
Summary: Six confidence intervals for the variance of a distribution are proposed. Extensive simulation study is performed to evaluate the performance of the intervals. A confidence interval based on an $$L$$-estimate of scale is found to be robust; in other words, regardless of the sample size and the distribution considered in the study, its actual probability of coverage is quite close to the specified confidence coefficient $$1-\alpha$$.

##### MSC:
 62F25 Parametric tolerance and confidence regions 62G15 Nonparametric tolerance and confidence regions 65C05 Monte Carlo methods
Full Text:
##### References:
 [1] Arnold B.C., A First Course in Order Statistics (1992) · Zbl 0850.62008 [2] Barham A.M., Unpublished Ph. D. Dissertation (1996) [3] Barnett V., Outliers in Statistical Data (1984) · Zbl 0638.62002 [4] Bickel Peter J., Mathematical Statistics: Basic Ideas and Selected Topics (1977) · Zbl 0403.62001 [5] DOI: 10.1111/j.1467-842X.1994.tb00637.x · Zbl 0825.62418 · doi:10.1111/j.1467-842X.1994.tb00637.x [6] DOI: 10.1080/03610927908827743 · doi:10.1080/03610927908827743 [7] DOI: 10.1214/aoms/1177705900 · Zbl 0093.15802 · doi:10.1214/aoms/1177705900 [8] DOI: 10.2307/1266226 · doi:10.2307/1266226 [9] DOI: 10.2307/2346782 · doi:10.2307/2346782 [10] DOI: 10.2307/2285324 · Zbl 0336.62036 · doi:10.2307/2285324 [11] DOI: 10.1214/aoms/1177703732 · Zbl 0136.39805 · doi:10.1214/aoms/1177703732 [12] DOI: 10.1214/aos/1176348783 · Zbl 0782.62039 · doi:10.1214/aos/1176348783 [13] Lehman E.L., Testing Statistical Hypotheses (1986) · doi:10.1007/978-1-4757-1923-9 [14] DOI: 10.2307/2289611 · doi:10.2307/2289611 [15] DOI: 10.1145/355606.361888 · Zbl 0244.65005 · doi:10.1145/355606.361888 [16] DOI: 10.1145/360827.360840 · Zbl 0273.65004 · doi:10.1145/360827.360840 [17] DOI: 10.2307/3002019 · doi:10.2307/3002019 [18] Tiku M.L., Robust Inference (1986) [19] Tukey J.W., Sankhyâ A. 25 pp 331– (1963)
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.