I'm new here on the forum, but I have a question about the possibility to combine the output of different t tests in one table.
I'll explain my question by using the auto dataset of stata.
I aslo read the answer on this question: https://www.statalist.org/forums/for...into-one-table
This answer gave the correct input to take my first step.
I used the following code in order to achieve a table with seven columns.
Code:
sysuse auto, clear ttest price if turn>=1 & rep78>2 , by(foreign ) loc mean1 = `r(mu_1)' loc mean2 = `r(mu_2)' loc se1 = `r(se)' loc se2 = `r(se)' loc obs1 = `r(N_1)' loc obs2 = `r(N_2)' * Write these with asdoc using the row option asdoc, save(output.doc) row(\i, Group 2, Group 1, \i, \i, \i, \i) title(Ttest) replace asdoc, save(output.doc) rowappend row(Price, `mean2', `mean1') dec(2) stars(all) asdoc, save(output.doc) rowappend row(\i, `se2', `se1') dec(2) stars(all) asdoc, save(output.doc) rowappend row(Observations, `obs2', `obs1') dec(0) stars(all)
Group 2 | Group 1 | Gear_ratio2 | Gear_ratio1 | Length 1 | Length 2 | |
Price | 6384.68 | 6164.19 | M2 | M1 | M2 | M1 |
775.71 | 775.71 | SE2 | SE1 | SE2 | SE1 | |
Observations | 22 | 42 | N | N | N | N |
Is this possible? And how can I achieve this?
Another question: is it possible to present the standard error for the first variable and the second variable instead of the overall standard error?
A last one: What is wrong with my code concerning significance stars?
Thanking you in advance,
0 Response to Combining results of different t tests in a table
Post a Comment