Dear Isabel,
Please mention the sample size. Do you mean non-normality of the
residuals or the variables? (treatment of both iassues are different).
If you have a sample size greater than 7000 observations than both
Shapiro -Wilk, Shapiro-Francia test are inappropriate. Try to use
graphical mehods instead (e.g. histogram, kdensity plots, Q-Q/P-P
plots, they give better results normally).
The question regarding how serious non-normality is, is a subjective
issue. If you feel that you can assume normality (CLRM) than it's up
to you. If you don't feel so, you can always choose ohter methods than
OLS.
Lastly, never forget to mention you problem in the "subject line" briefly.
M. Haider Hussain
Social Policy & Development Center
Karachi, Pakistan.
On 12/8/05, Isabel Vargas <[email protected]> wrote:
> Hello
>
> Please I have a econometric question not exactly about stata.
> I am working with cross-section data, and I use swilk sfrancia and iqr and I
> found that it reject normality, it is serious? If it is, what can I do?
>
> Thank you,
> Isabel
>
> _________________________________________________________________
> Charla con tus amigos en l=EDnea mediante MSN Messenger:
> http://messenger.latam.msn.com/
>
> *
> * 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/
