There is some literature on the sampling distribution of the sample
variance, where the parent distribution is highly non-normal,
particularly, this:
http://www.jstor.org/pss/3314982
The basic, chi2-based confidence interval can be calculated as below.
/***** Compute the sampling distribution of the sample variance *****/
clear*
sysuse auto
// assuming that the variable is normally distributed
scalar alpha = .05
sum price
di "The " 1-alpha " percent confidence interval for the sampling
variance: " r(Var) " is " ///
"[" ((`r(N)'-1)/invchi2(`r(N)'-1,1-alpha/2))*`r(Var)' ", "
((`r(N)'-1)/invchi2(`r(N)'-1,alpha/2))*`r(Var)' "]. "
/***** Compute the sampling distribution of the sample variance *****/
T
2010/2/1 Visintainer PhD, Paul <[email protected]>:
> Rosner's approach is based on the chi-square, but he does note that it is applicable only for normally-distributed samples.
>
> -p
>
> ___________________________________
> Paul F. Visintainer, PhD
> Director of Epidemiology and Biostatistics
> Baystate Medical Center
> Division of Academic Affairs - 3rd Floor
> 280 Chestnut Street
> Springfield, MA 01199
> (T) 413.794.7686
> (F) 413.794.7689
>
> -----Original Message-----
> From: [email protected] [mailto:[email protected]] On Behalf Of Lachenbruch, Peter
> Sent: Monday, February 01, 2010 11:36 AM
> To: '[email protected]'
> Subject: st: RE: RE: Fw: Calculating a confidence interval for population variance based upon sample.
>
> This is usually based on the chi-squared distribution - but it is rarely used because the variance is affected by a) outliers, b) distribution. I don't have my Rosner handy so I'm uncertain if it's the procedure he recommends.
> I would propose a couple of alternatives:
> 1. find a bootstrap confidence interval for the variance using percentiles or accelerated bootstrap.
> 2. Abandon the chi-square and get estimates of the percentiles of interest (e.g., 0.025, 0.05, 0.1 and their complements) with centile. This may have a problem if sample size is small.
> After finding the confidence interval for the variance, you can find the standard deviation easily enough
>
> Tony
>
> Peter A. Lachenbruch
> Department of Public Health
> Oregon State University
> Corvallis, OR 97330
> Phone: 541-737-3832
> FAX: 541-737-4001
>
>
> -----Original Message-----
> From: [email protected] [mailto:[email protected]] On Behalf Of Visintainer PhD, Paul
> Sent: Monday, February 01, 2010 7:15 AM
> To: '[email protected]'
> Subject: st: RE: Fw: Calculating a confidence interval for population variance based upon sample.
>
> Carl,
>
> There is a procedure in Rosner [Rosner, B. "Fundamentals of Biostatistics, (6th ed.), Duxbury, Press (The Thompson Company): Belmont, CA, 2006, pgs. 199-201.]
>
> If you send me your email, I can pass along a "not-ready-for-prime-time" .ado that I use for the calculation.
>
> -p
>
>
>
> ___________________________________
> Paul F. Visintainer, PhD
> Baystate Medical Center
> Springfield, MA 01199
>
>
> -----Original Message-----
> From: [email protected] [mailto:[email protected]] On Behalf Of Carl Mastropaolo
> Sent: Monday, February 01, 2010 9:47 AM
> To: [email protected]
> Subject: st: Fw: Calculating a confidence interval for population variance based upon sample.
>
> Hi
>
> I have STATA10.
>
> I can not figure out how to calculate a simple confidence interval for a
> population variance
> based upon sample data. STATA will calculate the sample standard deviation,
> but I can not figure out how to compute a confidence interval for the
> population standard deviation. Is there a command which will do this?
>
> Any help?
>
> Thanx,
> Carl
>
> *
> * For searches and help try:
> * http://www.stata.com/help.cgi?search
> * http://www.stata.com/support/statalist/faq
> * http://www.ats.ucla.edu/stat/stata/
>
> ----------------------------------------------------------------------
> CONFIDENTIALITY NOTICE: This email communication and any attachments may contain confidential and privileged information for the use of the designated recipients named above. If you are not the intended recipient, you are hereby notified that you have received this communication in error and that any review, disclosure, dissemination, distribution or copying of it or its contents is prohibited. If you have received this communication in error, please reply to the sender immediately or by telephone at (413) 794-0000 and destroy all copies of this communication and any attachments. For further information regarding Baystate Health's privacy policy, please visit our Internet web site at http://www.baystatehealth.com.
>
> *
> * For searches and help try:
> * http://www.stata.com/help.cgi?search
> * http://www.stata.com/support/statalist/faq
> * http://www.ats.ucla.edu/stat/stata/
>
> *
> * For searches and help try:
> * http://www.stata.com/help.cgi?search
> * http://www.stata.com/support/statalist/faq
> * http://www.ats.ucla.edu/stat/stata/
>
> *
> * For searches and help try:
> * http://www.stata.com/help.cgi?search
> * http://www.stata.com/support/statalist/faq
> * http://www.ats.ucla.edu/stat/stata/
>
--
To every ω-consistent recursive class κ of formulae there correspond
recursive class signs r, such that neither v Gen r nor Neg(v Gen r)
belongs to Flg(κ) (where v is the free variable of r).
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
