Interaction term of a global variable list with a factor variable

[Aakanksha Melkani]

Hi,
I am using Stata 14.0 and trying to obtain full interaction of my variable list with a factor variable. Is there an easy way Stata allows me to do this while using global variable lists?

Here is what I tried:

global lexppr lexpretprice lexpfraprice /// Expected output prices
global lcurrpr lretprice lfraprice /// Current output prices
probit mzmktpos i.lc##c.($lexppr $lcurrpr) i.lc##i.(malehead year)

Stata ran the regression but ignored the interaction of factor variable "lc" with my global variables : i.lc##c.($lexppr $lcurrpr). It did not give me an error message either.

Thanks for any help in advance,
Aakanksha

[Clyde Schechter]

If this is the code that you actually ran, Stata did not run the regression, nor did it execute your -global lucurrpr- command. The reason is that you used the /// at the end of your commands. The /// token tells the Stata parser to a) treat whatever follows on the same line as comment, and b) treat the next line as a continuation of the command. Consequently Stata interprets the three lines of code you show as a single command defining a global macro that starts with lexpretprice and ends with i.lc##I.(malehead year).

To get these to be three separate commands, and still preserve your comments, change the ///'s to //. // tells the parser to treat the rest of the line as a comment, but it does not treat the next line as a continuation of the command.

[Aakanksha Melkani]

Thanks much for the help Clyde! It works perfectly now. It was indeed a silly mistake on my part.