Testing the variance of skewed distributions

Lee, S. J.; Sa, Ping

doi:10.1080/03610919808813509

Cited by 2 publications

(5 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, the Type I error rates become more or less uncontrollable when either the alpha level gets small or the sample size is reduced. These results confirmed the recommendations of Lee and Sa (1998) that Zs is more suitable for moderate to large sample sizes and alpha levels not too small. Although Zs was specifically designed for the skewed distributions, it actually works reasonably well for the heavy-tailed distributions as long as the sample size and/or the alpha level are not too small.…”

Section: Type I Error Comparisonsupporting

confidence: 79%

“…Another alternative is presented in Kendall (1994) and Lee and Sa (1998). The robust chi-square statistic χ test with skewed distributions.…”

Section: Introductionmentioning

confidence: 99%

“…Lee and Sa (1998) performed another study on a right-tailed test of variance for skewed distributions. A method similar to the previously proposed study was employed with the primary difference being in the estimated coefficient of skewness, 1 β .…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Right-tailed Testing of Variance for Non-Normal Distributions

Long

2005

J. Mod. App. Stat. Meth.

Self Cite

View full text Add to dashboard Cite

show abstract

Section: Type I Error Comparisonsupporting

confidence: 79%

“…Another alternative is presented in Kendall (1994) and Lee and Sa (1998). The robust chi-square statistic χ test with skewed distributions.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Right-tailed Testing of Variance for Non-Normal Distributions

Long

2005

J. Mod. App. Stat. Meth.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Another alternative is presented in Kendall (1994) and Lee and Sa (1998 Lee and Sa (1996) derived a new method for a right-tailed variance test of symmetric heavy-tailed distributions using an Edgeworth expansion (see Bickel & Doksum, 1977), and an inversion type of Edgeworth expansion provided by Hall (1983),…”

Section: Introductionmentioning

confidence: 99%

“…The bootstrap requires extensive computer calculations and some programming ability by the practitioner making the method infeasible for some people. Although the jackknife method is easier to implement, it is a linear approximation to the bootstrap method and can give poor results when the statistic estimate is nonlinear.Another alternative is presented in Kendall (1994) and Lee and Sa (1998 where θˆ is any statistic, and θ , ) (θ σ where After a simulation study, their study found their test provided a "controlled Type I error rate as well as good power performance when sample size is moderate or large" (p. 51). Lee and Sa (1998) performed another study on a right-tailed test of variance for skewed distributions.…”

mentioning

confidence: 99%

Vol. 4, No. 1 (Full Issue)

Editors¹

2005

J. Mod. App. Stat. Meth.

View full text Add to dashboard Cite

ScopeThe Directory aims to be comprehensive and cover all open access scientifi c and scholarly journals that use a quality control system to guarantee the content. All subject areas and languages will be covered. The easy way to fi nd open access journalsThe Directory of Open Access Journals covers free, full text, quality controlled scientifi c and scholarly journals. It aims to cover all subjects and languages. DIRECTORY OF OPEN ACCESS JOURNALS JMASM Algorithms and Code -311Hakan Demirtas JMASM16: Pseudo-Random Number Generation in R for Some Univariate Distributions (R) May, 2005, Vol. 4, No. 1, 2-352 1538 vii JMASM is an independent print and electronic journal (http://tbf.coe.wayne.edu/jmasm) designed to provide an outlet for the scholarly works of applied nonparametric or parametric statisticians, data analysts, researchers, classical or modern psychometricians, quantitative or qualitative evaluators, and methodologists. Work appearing in Regular Articles, Brief Reports, and Early Scholars are externally peer reviewed, with input from the Editorial Board; in Statistical Software Applications and Review and JMASM Algorithms and Code are internally reviewed by the Editorial Board.Three areas are appropriate for JMASM: (1) development or study of new statistical tests or procedures, or the comparison of existing statistical tests or procedures, using computerintensive Monte Carlo, bootstrap, jackknife, or resampling methods, (2) development or study of nonparametric, robust, permutation, exact, and approximate randomization methods, and (3) applications of computer programming, preferably in Fortran (all other programming environments are welcome), related to statistical algorithms, pseudo-random number generators, simulation techniques, and self-contained executable code to carry out new or interesting statistical methods. Elegant derivations, as well as articles with no take-home message to practitioners, have low priority. Articles based on Monte Carlo (and other computer-intensive) methods designed to evaluate new or existing techniques or practices, particularly as they relate to novel applications of modern methods to everyday data analysis problems, have high priority.Problems may arise from applied statistics and data analysis; experimental and nonexperimental research design; psychometry, testing, and measurement; and quantitative or qualitative evaluation. They should relate to the social and behavioral sciences, especially education and psychology. Applications from other traditions, such as actuarial statistics, biometrics or biostatistics, chemometrics, econometrics, environmetrics, jurimetrics, quality control, and sociometrics are welcome. Applied methods from other disciplines (e.g., astronomy, business, engineering, genetics, logic, nursing, marketing, medicine, oceanography, pharmacy, physics, political science) are acceptable if the demonstration holds promise for the social and behavioral sciences. This article considers a J by K ANOVA design where all JK groups are dependent and where...

show abstract

Testing the variance of skewed distributions

Cited by 2 publications

References 3 publications

Right-tailed Testing of Variance for Non-Normal Distributions

Right-tailed Testing of Variance for Non-Normal Distributions

Vol. 4, No. 1 (Full Issue)

Contact Info

Product

Resources

About