Different coefficients on the same independent variables in regression analysis based on dummy variables

[John Charlse Doe]

Dear community members,

I would like to make the table on the left side of the picture below (mine is on right). It has three different coefficients on the same independent variables based on three dummy variables. However, whenever I create the table, using outreg2, only one column was printed out. Would there be a solution to this?

[Carlo Lazzaro]

John:
take a look at the wonderful -table- suite of commands in Stata 17.

[Hemanshu Kumar]

The very first example in

help outreg2

(see the commands under Example 0) creates a table with two columns, one for each model. You should compare with that and see what you might be doing wrong. For instance, are you specifying the -replace- option, so that the version of your file with the first model is getting overwritten? You may need to change that.

For others: outreg2 is a community-contributed command; see

net describe outreg2, from(http://fmwww.bc.edu/RePEc/bocode/o)

[John Charlse Doe]

Dear all, appreciate your contributions to my thread. The table on the left was based on the result of one single regression analysis, which is why all three columns shared the same Adj.R^2. Somehow, the authors allowed the coefficients to vary by bank assets in a single (not separate three regressions) regression, using a "reghdfe" package. I think they added two indicator variables in the independent variable part and printed out the result as above. FYI, I have no problem with three accumulated regression results from three separate regressions based on sub-samples. Sorry for the confusion.