Friedrich,
my initial experience with -mfx- taking more than 39 hours after
a -herprob- estimation was based on a model with 44 regressors and 87 000
observations, CPU speed 1,92 Ghz, and the following version of Stata 8:
. update query
Stata executable
folder: D:\Programme\Stata8\
name of file: wstata.exe
currently installed: 23 Jan 2003
latest available: 10 Jan 2005
Indeed, it does not look very updated. I am about to update it and I'll
run the model again afterwards to see if there is a difference.
Thomas
--------- Friedrich H�bler wrote:
I would like to hear more about your experience with -mfx-. When you
say that the large model had not finished after 39 hours, are you
talking about Stata 8.0 or Stata 8.2 with all the latest updates? I
recall that the speed of -mfx- was increased significantly after the
initial release of Stata 8.
--- Thomas Corneli�en wrote:
I have made the experience that -mfx- has been considerably speeded
up in Stata 9.
My experience relies on -mfx- after the heteroscedastic probit
model -hetprob-. For example, in one model Stata 8 took 3:24 hours
for -mfx
compute- after -hetprob-, while Stata 9 took 1 minute 24 seconds. A
larger
model hadn't finished after 39 hours in Stata 8, but it took only
10 minutes in Stata 9.