ttest by different classification standard

[dara wang]

Hi,

I have a bunch of variables, say: a,b,c,d,e. I want to do ttest by different classification. The classification variable is: aa, bb, cc, dd, ee

I understand that I can use ttable to generate a beautiful table of t test between two groups of many variables. Is there any way to generate a table with only one variable but by different classification standard, like: ttable a, by(aa, bb, cc, dd, ee)

(Stata does not allow more than one variable in by statement)

I appreciate it very much if someone can tell me some new functions or any other way to get what I want.

Thanks!

[Carlo Lazzaro]

Dara:
welcome to this forum.
I'm not sure I got what you're after.
Anyway, I suspect that you should consider -regress- or -anova- instead of -ttest-.

[Marcos Almeida]

If I understood right, you can have a loop to provide the estimations. But the results would be presented in different tables.

[Rich Goldstein]

not sure I completely understand; but, as -ttable- is available from SSC (as the FAQ asks you to explain), SSC also contains -ttable2- and -ttestable-, either of which might do what you want

[dara wang]

Marcos Almeida Yes, I can have a loop to do what I want. But it will show in different tables. I want them to show in the same table.