[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
RE: st:problems with outliers
> First, I'm trying to follow your suggestion about
> using the command >if but I have 12 variables: do you
> think I could use it the same?(sorry, if pheraphs my
> question it's a bit silly: but isn't this method the
> same as dropping observations? I mean Sytata
> automatically won't take care of the observations
> which hhavve variables outside the range?)
There is no Stata reason why a complicated -if- condition
should not be used. It is not the same as dropping
I don't understand the last question.
> Second, I must realize a regression using a logit
> model: what's the equivalent command as qreg I could
> use in that case?
I don't think there is one, at least not in Stata.
> Third, I'd like to try also with the method of
> transformation: could you give me some suggestion
> about what transformstion should I perform?
Very difficult to advise without seeing your
data or knowing your project. I presume from
above that the outliers are all in the covariates.
If any transformations are appropriate, they
would be of the covariates.
Check out -hadimvo- with your covariates.
* For searches and help try: