Bonett [1] provides an approximate confidence interval for s and shows it to be nearly exact under normality and, for small samples, under moderate non-normality. This paper proposes a new method for determining the confidence interval for . We follow the suggestion of Bonett [1] and include any prior kurtosis information. An important modification from the Bonett method is to base the interval on , the ( /2)100th quantile of Student T distribution, rather than on , the ( /2)100th quantile of standard normal distribution. Further, the use of the median as an estimate of the population mean gives a slightly higher coverage probability for the confidence interval for when data are from skewed leptokurtic distributions. Monte Carlo Simulation results for selected normal and non-normal distributions show that the confidence intervals obtained from the new method have appreciably higher coverage probabilities than the confidence intervals from the original Bonett method that does not use prior kurtosis information, and also higher coverage probabilities than the Bonett method that does use prior kurtosis information.
[1]
Ulf Olsson,et al.
Confidence Intervals for the Mean of a Log-Normal Distribution
,
2005
.
[2]
Franklin A. Graybill,et al.
Introduction to the Theory of Statistics, 3rd ed.
,
1974
.
[3]
Wei Pan,et al.
Approximate confidence intervals for one proportion and difference of two proportions
,
2002
.
[4]
Douglas G. Bonett,et al.
Approximate confidence interval for standard deviation of nonnormal distributions
,
2006,
Comput. Stat. Data Anal..
[5]
Paul Kabaila,et al.
VALID CONFIDENCE INTERVALS IN REGRESSION AFTER VARIABLE SELECTION
,
1998,
Econometric Theory.
[6]
Lewis H. Shoemaker,et al.
Fixing the F Test for Equal Variances
,
2003
.