Thanks Maarten and Kit. Scaling the regressor will do the trick and
solves my problem.
My general preference is to change reported estimates rather than the
underlying data. Multiplying by 100 seems like an easy task,
especially since -estout- can transform coefficients (and presumably
se. if I could figure out how). Of course I am willing to compromise
that principle to get my problem fixed -- thanks for your suggestions!
Maarten, in my context the linear probability model fits quite well.
Your point is well taken and in fact my logit marginal effects are
very similar. For better or worse economists tend to like LPM, see
e.g. page 107 in Angrist and Pischke* for a common view.
Cheers,
Bert
PS Apologies for not specifying: as you guessed my -estout- is from
SSC. I use Stata/SE 10.1
* Angrist and Pischke (2009) "Mostly Harmless Econometrics",
http://books.google.com/books?id=bUZzyXeV1JUC
On Mon, Feb 1, 2010 at 7:52 AM, Kit Baum <[email protected]> wrote:
> <>
>
> As you're using a LPM, why don't you just scale the regressors /100?
>
> On Feb 1, 2010, at 2:33 AM, Bert wrote:
>
>> I am reporting coefficients from a linear probability model using
>> - -estout- and -esttab-. I would like to multiply the coefficients with
>> 100 so that readers can interpret them as percentage point change. I
>> tried the -transform(@*100)- option in estout which works for the
>> coefficients but struggle with the standard errors. Any suggestions
>> greatly appreciated.
>
>
> Kit Baum | Boston College Economics & DIW Berlin | http://ideas.repec.org/e/pba1.html
> An Introduction to Stata Programming | http://www.stata-press.com/books/isp.html
> An Introduction to Modern Econometrics Using Stata | http://www.stata-press.com/books/imeus.html
>
>
> *
> * 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/
>
*
* 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/
