Korn, E. and Graubard, B. "Predictive Margins for Survey Data", Biometrics,
about June 1999. You can ignore the survey design issues if they don't
apply to you, and the rest still holds. BTW, they are only the same for
mean based models, like OLS, and the s.e. are different.
Bryan Sayer
Statistician, SSS Inc.
[email protected]
(p.s. I might not have the title just right, but it's close)
-----Original Message-----
From: [email protected] [mailto:[email protected]]
Sent: Wednesday, October 02, 2002 3:20 PM
To: [email protected]
Subject: st: Marginal effects: at what value should you calculate?
In "Econometric Analysis" by William H. Greene, 3rd edition. He
states on page 876: "For computing marginal effects, one can
evaluate the the expressions at the sample means of the data or
evaluate the marginal effects at every observation and use the sample
average fo the individual marginal effects. The functions are
continuous, so Theorem 4.3 (the Slutsky theorem) applies; in large
samples these will give the same answer. But that is not so in small or
moderate sized samples. Current practice favors averaging the
individual marginal effects when it is possible to do so."
Is there any more definitive guidance out there for the values at which
one should calculate marginal effects? While I agree that it makes
intuitive sense to do the calculations at every observation and
average, is there a theoretically sound reason to do so?
Sincerely,
Eric A. Powers
Assistant Professor of Finance
The Moore School of Business
University of South Carolina
Columbia SC, 29208
---
[This E-mail scanned for viruses by Declude Virus]
*
* 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/
*
* 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/
