Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

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

st: Error message

From   "Fred Dzanku" <[email protected]>
To   <[email protected]>
Subject   st: Error message
Date   Tue, 12 Oct 2010 00:49:37 +0100

Dear All, 
I estimated the following SUR model:
	qui: bootstrap, rep(20) cl(clust) str(dist) seed(10031976)  ///
		nodots: sureg (pgap_h $flbpr) (psev_h $flbpr)

after which I requested for the following:
	margins, eyex(*) predict(equation(pgap_h))

which worked fine but when I tried this for the second equation I get an
error message which reads as follows:
	"could not calculate numerical derivatives -- discontinuous region
with missing values encountered
This happens whether or not I use the bootstrap option. It also happens with
mvreg. This however does not occur when I estimate the equations
individuation. I tried introducing the variables one after the other to see
if some particular variable(s) might be responsible and found that out of
the 24 variables contained in $flbpr 14 of them work fine but when I
introduce any of the other 10 then the error message comes up. I've check
these variables and don't seem to find exactly what the problem is. Can
anyone help please? 

*   For searches and help try:

© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index