I don't understand the numbers in your example:
"
y_pred=-2.33-0.431*x (x being significant)
Turning off the effect of X thus gives me: y_pred=-2.33-(0.431*0) and
Pr(z<2.33)=0.99%
Tuning on the effect: y_pred=-2.33-(0.431*1)=-2.761 and
Pr(z<2.761)=0.29%
"
Whether x is significant or not makes no difference. If you want the
marginal effect of a discrete change for a regression model showing
y_pred=-2.33-0.431*x
would you not show
di normal(-2.33-0.431)-normal(-2.33)
which is -.00702184 or a 0.7 percentage point reduction from about
4952 expected to have positive outcomes in a sample of 5e5 to about
1441 cases?
di round(5e5*normal(-2.33))
4952
di round(5e5*normal(-2.33-0.431))
1441
di round(5e5*(normal(-2.33-0.431)-normal(-2.33)))
-3511
But with a single regressor, you could get this from
tab y x
more easily!
With additional regressors, there are more options for calculating
marginal effects.
On Jan 11, 2008 11:57 AM, <[email protected]> wrote:
> Thanks again for helping me with the logic of these calculations.
> It turned on to be a bit on the side of a typical Stata topic, sorry about that.
> Alex
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
