My thanks to Nick Winter, and pardon me for mis-typing your name. I have no missing values at all, I felt with such a small size sample (Obs=26) I could do better by presenting the conventional 95% C.I. for the mean-difference and a bootstrapped (bias-corrected) one.
You are absolutely right, I could have used "summarize" or "ci" for that matter.
Many thanks for your help.
Amani
-----Original Message-----
From: Nick Winter [mailto:[email protected]]
Sent: 12 June 2002 18:59
To: [email protected]
Subject: st: RE: RE: RE: RE: Bootstrapping and t-tests
> Ok, thanks to Nick Winters, I have learned something new
By the way, it is "Winter" -- no "s" -- a pet peeve of mine.
> Warning: Since ttest is not an estimation command or does
> not set e(sample), bs has no way to determine
> which observations are used in calculating the
> statistics and so assumes that all observations
> are used. This means no observations will be
> excluded from the resampling due to missing values
> or other reasons.
> If the assumption is not true, press Break, save
> the data, and drop the observations that are to
> be excluded. Be sure the dataset in memory
> contains only the relevant data.
On the substance, the warning means that you must ensure there are no
missing observations on your diff variable. You can do this by typing
. count if diff==.
If there are any, you should drop those observations before running your
bootstrap:
. preserve
. drop if diff==.
<<run bootstrap>>
. restore
By the way, you could just use the -summarize- command, rather than the
-ttest-, if all you want is the mean.
Nick Winter
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
