Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: Questions on marginal effects for the model including squared term


From   Xi Pan <mpanxi@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   st: Questions on marginal effects for the model including squared term
Date   Sat, 13 Feb 2010 15:21:34 -0600

Dear All,
I am using Stata 9.2.
I run mlogit model and then use mfx to obtain the marginal effect for
each category. In the model, the dependent/categorical variable
contains 4 categories. But covariates consist of ‘years of work
experience’, its squared term and other variables such as gender.
After I run mfx, I obtain the m_experience, m_experience-squared-term
and m_gender and the corresponding standard errors in each category.

My question is regarding the true marginal effect of experience. For example,

The true marginal effect of experience in category 1 = _m[experience]
+ 2*_m[experience squared term]*(average of years of experience). So,
in theory, we are supposed to obtain one standard error for this term.

My questions are
1. How to calculate the standard error for this true marginal effect
for experience?
2. In the above equation, does stata use ‘average of years of
experience’  in category 1, or use the average in the full sample
(having all 4 categories) ?
 3. After I use mfx in each category, how to save variance-covariance
matrix of marginal effects?

This is the final step to revise my dissertation, so any information
from you would help me to move on.

Thanks a lot!

Sincerely,

Xi

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index