Double sorted portfolio - Error Estimation

[Bella Moirie]

I have to make a double sorted portfolio analysis.
First I need to regress a simple linear model with 4 independent variables, and find the error term.
I need to do this for around 100 persons in a sample, in a quarterly time span for 6 years.

From the error term I should then estimate the standard deviation.

Then I need to categorize the standard deviations of the error terms into four portfolios, from low to high.
How could I estimate the error for all persons at every different time all at once?

The data that I use is in Excel and looks like this:
Dates Name Variable1 Variable2 ... etc
Q12013 X 100
Q22013 X 103
Q32013 X 122
Q42013 X 98
Q12014 X 128
.......
.....
Q12013 Y 110
Q22013 Y 105
Q32013 Y 132
Q42013 Y 39
Q12014 Y 138
.......
.....
Q12013 Z 87
Q22013 Z 115
Q32013 Z 139
Q42013 Z 87
Q12014 Z 132

[William Lisowski]

Welcome to Statalist.

You have accidentally posted your topic in Statalist's Mata Forum, which is used for discussions of Stata's Mata language, which is different than Stata's command language that you will be using. Your question will see a much larger audience if you post it in Statalist's General Forum.

Also, if you have not already done so, take a look at the Statalist FAQ linked to at the top of this page for posting guidelines and suggestions.