For heteroskedasticity, use robust standard errors.
For normality, try to transform your variable (log, square root) and to eliminate outliers. Command -ladder- searches a subset of the ladder of powers (cubic, square, square-root, log, reciprocal root, reciprocal, reciprocal square, reciprocal cubic) for a transform that converts a variable into a normally distributed one. Which command did you use to test for normality?
Nicola
At 02.33 11/05/2007 -0400, gargouri yosra wrote:
>I have used a regression panel data with random effects. the tests shows th=
>e presence of heteroskedasticity problem and the test for normality reject=
> H0.
>what I can do?
>thank you
*
* 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/
