Quantcast
Channel: Statalist
Viewing all articles
Browse latest Browse all 65593

Why nlcom standard deviation for exponentiated estimators differs from the manually calculated one ?

$
0
0
Hi Statalisters,

The marginal effect of my variable of interest subsequent to a panel fixed effects estimator is given by
Array

I am using command nlcom in Stata 14.0 to obtain the standard deviation of the per capita income when the effect is zero with
Code:
nlcom exp(-1*(_b[est1]/_b[est2]))
As this follows log normal distribution, to calculate the variance manually I use
Array .

The standard deviation I calculate using this formula is different from the one Stata yields. Also Stata does not correct for the mean coefficient in accordance with Array , but gives Array instead.


Could anybody kindly explain why this happens?

Thanks

Subhani

Viewing all articles
Browse latest Browse all 65593


<script src="https://jsc.adskeeper.com/r/s/rssing.com.1596347.js" async> </script>