|
Size: 1369
Comment:
|
Size: 1377
Comment:
|
| Deletions are marked like this. | Additions are marked like this. |
| Line 53: | Line 53: |
| ---- |
Stata Regress
The regress command runs a regression model.
Contents
Usage
regress dependent independent
Factor Variables
To regress on the levels of a variable rather than its numeric value, prefix the variable name with i..
To regress on an interaction of variables, delimit the two variable names with #. Or use ## to indicate a full factorial (both variables and the interactions).
To create an interaction with a continuous variable, prefix them with c..
Estimates
To store the estimation results, try:
estimates store example
To display the estimation results, try:
estimates table example estimates table example, b(%10.3f) star
The star option shows significance.
While coefficients are always displayed by estimates table, they can optionally be formatted with the b(%format) option (as above). There are several optional statistics available as well:
se or se(%format) for standard errors
t or t(%format) for t- or z-statistics
p or p(%format) for p-values
See also
Stata manual for logistic post-estimation