Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


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

st: Problem while generating limited predictor after spreg


From   rraciborski@stata.com (Rafal Raciborski, StataCorp)
To   statalist@hsphsun2.harvard.edu
Subject   st: Problem while generating limited predictor after spreg
Date   Wed, 28 Nov 2012 14:32:43 -0600

Lin-Han <chianghlh@gatech.edu> has a question involving the calculation of
predicted values after the user-written -spreg- command:

> I ran into a problem while generating limited predictor after spreg...
> I have a dataset with 47,000 observations, and ... when I run this command
> after spreg:
>
> predict yhat, limited
>
> The server ran for more than 60 hours before I forced it to stop, and
> no result is yet generated.

The formulas for spatial predictors are described in section 6.3 of the
working paper by Drukker, Prucha, and Raciborski (2011).  The limited
information predictor involves several matrix-inverse operations, each running
in O(n^3) time.  For n=47,000, my rough calculation puts each matrix-inverse
operation at 67 hours.

-- Rafal
rraciborski@stata.com

References
----------
Drukker D M, Prucha I R, and Raciborski R. 2011.  Maximum-likelihood
and generalized spatial two-stage least-squares estimators
for a spatial-autoregressive model with spatial-autoregressive disturbances.
Working paper, University of Maryland, Department of Economics, 
http://econweb.umd.edu/~prucha/Papers/WP_spreg_2011.pdf.
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


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