[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: SV: SV: st: From probit to dprobit to interpretation

From   "Austin Nichols" <[email protected]>
To   [email protected]
Subject   Re: SV: SV: st: From probit to dprobit to interpretation
Date   Fri, 11 Jan 2008 13:28:32 -0500

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
 Tuning on the effect: y_pred=-2.33-(0.431*1)=-2.761 and
Whether x is significant or not makes no difference.  If you want the
marginal effect of a discrete change for a regression model showing
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))
di round(5e5*normal(-2.33-0.431))
di round(5e5*(normal(-2.33-0.431)-normal(-2.33)))

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:

© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index