renaming regressors when presenting the results?

[Juan Price Elton]

In order to run a regression I named the regressors as x1 x2 x3 x3*x1 and so on (I have around 12 regressors) and when doing the "estimate table" to present the results I would like to have the original regressors' names, let's say x1= educ x2=exp etc.... is there anyway of doing this? Many thanks

[Marcos Almeida]

You can use - label variable - so as to, well, label the variables. Then, under - estimates table - , use the - varlable option. Hopefully that helps.

[William Lisowski]

I would be interested in learning why you needed to rename the regressors in order to run your regression. That is not commonly an issue, certainly not one I recall seeing on Statalist, so I'm interested in knowing how it might arise. I notice in an earlier topic you indicated you were using "SFA (parametric)" - by that do you mean you are using Stata's frontier command for Stochastic Frontier Analysis?

But in any event, Marcos's advice will likely take you from where you are now to where you want to be.

[Juan Price Elton]

hi Marcos : thank you very much !!! hi William: The thing is that in order to create a translog input distance function I use the original regressors to create many new regressors, some of which are normalised by one of the inputs. that is a ver long function, with many interactions.. to build all those terms I use many loops, therefore I go like "foreach i....." being each I the subscript of an x and I have maaaany x variables... I don't want to present the output in terms of those variables. I want it to be read in terms of the original variables so the reader knows what each variable is. Thanks again. Kind regards

[Phil Bromiley]

Sometimes it is easier just to set up the report table (in word or whatever) and then generate the results table (estout, outreg2 or whatever), and cut-and-paste into the report table. That way you can have anything you want in the variable name in the report including subscripts etc. without messing with Stata.

[Marcos Almeida]

I confess I sometimes fiddle with these programs so as to produce neat tabs, but, most of the time, I prefer to costumize a table the old way I mean, just copying and pasting the chunks I do wish to present. The more time I spend doing this procedure, the more intimacy I get with the data.

[Juan Price Elton]

many thanks !!! All the best