Dear All,
I am working with firm level data. I want to generate a variable lsales = ln(sales) and want to replace the values in lsales as missing if the values in the variable lsales are higher than mean+3*standard deviation. While doing so, I want the summary stat of lsales by grouping state and sector and if the observation of lsales is more than 1000 in the individual group (state-sector) then find the outlier and replace it as missing. If the number of observation is less than 1000 in any of the group (state-sector) then I want to group lsales by the state only and find out the summary stat and replace the outlier with missing values. Please suggest to me how should I code this, as I am unable to visualise the loop for the condition.

Thanks
Prakash