It might be worth having a look at Nick Cox's -winsor- command, eg:Code:ssc describe winsor +/- ssc install winsor
Sir,
i am new at using stata. I want to winsorized at the 1% and 99% levels in order to minimize the effect of outliers.
I have many variables. like
y = a + b1 X1 + b2 X2 + b3X3 +b4X4 +....+b12X12 + error
where b is beta
Please guide me is there any way that i can remove outliers.
Regards
S.Franck
It might be worth having a look at Nick Cox's -winsor- command, eg:Code:ssc describe winsor +/- ssc install winsor
The extremes command of Nick Cox shows the bottom 5 and top 5 outliers. I had many observations and I wanted to check the observations first to see if there were any data problems. Maybe it can be useful for you too.
|
|