Dear Jan,
I run the demo code in your GitHub:

Partial out firm and year fixed effects (requires reghdfe):
nwxtregress cap_cons compensation net_surplus , dvarlag(Wt,mata timesparse) ivarlag(Wt: compensation,mata timesparse ) seed(1234) absorb(ID Year)

I encounter the following error message:
reghdfe Mata library not found or error in reghdfe.
r(199);

I use stata 17MP and update the reghdfe.
Thank you!
Peng

nwxtregress cap_cons compensation net_surplus , dvarlag(WSpmat) seed(1234)
error:
Traceback (most recent call last):
File "", line 1, in
ModuleNotFoundError: No module named 'scipy'
(43 lines skipped)

when I run this code, I met this error.

i want to output the log marginal likelihood,could the command do this?

Hello,
while trying to use estat imapct after successfully run nwxtregress, Stata gives the following error message, wondering if it is a bug.

. estat impact
----+--- 10 --+--- 20 --+--- 30 --+--- 40 --+--- 50 %
J(): 3900 unable to allocate real [2147379216,50]
SparseMultiply(): - function returned error
BarryPace(): - function returned error
calc_DIEffects(): - function returned error
: - function returned error
r(3900);

Stata 16.1

Dear Jan,

It seems that multiple matrices could not be added to dvarlag(), but only multiple matrices in ivarlag(). See if you could post the solution to it.
Thank you very much!

Best,
Wenna

Hi, Jan.

Happy new year!
When running the code, the following errors return in the 0.13 version of nwxtregress.
Is it anything suspicious about the data that cause the issue?

Thank you,
Wenna

When I run the example regression, Stata shows "This is an alpha version! Results may change. ".

System: Windows 10
Stata Version: 15.0

Dear Jan,

I came across the following errors. The error code is 3301.
I checked the id identifiers in the spatial network, which match across.
Wdistance is the spatial distance of cities.
Wdistancetry is the truncated version of Wdistance.

The same code worked perfectly for another dataset.

See if you have any solutions or insights for this issue.
Thank you very much!

Best,
Wenna

Dear Jan,

I found it's feasible to add "fe" to the end of nwxtregress.
Does it mean adding fixed effect to the model, same as adding "fe" to xtreg when running the panel data?

best,
Wenna

Display details, change se name

Hi,
When I run nwxtregress cap_cons compensation net_surplus , dvarlag(WSpmat) seed(1234) following the codes in help page, the error information is as follows:
'''
Initialise Grid (200)
----+--- 10 --+--- 20 --+--- 30 --+--- 40 --+--- 50 %
.................................................. 50
.................................................. 100
Griddy Gibbs (2000)
----+--- 10 --+--- 20 --+--- 30 --+--- 40 --+--- 50 %
mm_colrunsum(): 3001 expected 1 arguments but received 3
GriddyGibbsi(): - function returned error
GriddyGibbs(): - function returned error
estimatesar(): - function returned error
: - function returned error
r(3001);
'''
My Stata is SE 17.0 for Windows.

Thanks and best wishes.

Dear Jan,
I run the demo code in your github , eg VA data set to estimate a SAR mode, the estimation process goes smoothly but while I try to get the imact, estat impact, it returns the following errors:
estat impact
----+--- 10 --+--- 20 --+--- 30 --+--- 40 --+--- 50 %
*: 3200 conformability error
SparseMultiply(): - function returned error
BarryPace(): - function returned error
calc_DIEffects(): - function returned error
: - function returned error

As you methioned in one of the discussion, I need to install the new version of moremata library in Stata. The problem is that even I install it, the errors are still there, I sincerely hope that you can help me.

the coding is just like this :

use IO
keep if Year == 1998
drop buyer_industry seller_industry
replace sam = 0 if sam < 0
replace sam = 0 if ID1==ID2
reshape wide sam, i(ID1) j(ID2)
spset ID1
spmatrix fromdata WSpmat = sam* , replace
clear
use VA
nwxtregress cap_cons compensation net_surplus , dvarlag(WSpmat) seed(1234)
estat impact

Thank you very much!

Originally posted by @zhangzhqnk in #5