Near singular matrix, PMG

For technical questions regarding estimation of single equations, systems, VARs, Factor analysis and State Space Models in EViews. General econometric questions and advice should go in the Econometric Discussions forum.

Moderators: EViews Gareth, EViews Moderator

aawoad
Posts: 8
Joined: Thu Dec 24, 2020 1:49 am

Near singular matrix, PMG

Postby aawoad » Thu Dec 24, 2020 1:55 am

Dear All
Greeting
I'm doing an estimation using PMG, my panel structure is T(2000-2019, N=78). I don't have any dummy variables, when u run PMG, I received "Near singular matrix". When I tried in STATA is working. What do you suggest?

startz
Non-normality and collinearity are NOT problems!
Posts: 3775
Joined: Wed Sep 17, 2008 2:25 pm

Re: Near singular matrix, PMG

Postby startz » Thu Dec 24, 2020 7:16 am

In general, Stata silently drops variables when there is perfect multicollinearity while EViews gives a warning message instead. You might post your EViews file including the equation to be estimated.

aawoad
Posts: 8
Joined: Thu Dec 24, 2020 1:49 am

Re: Near singular matrix, PMG

Postby aawoad » Mon Dec 28, 2020 1:27 am

Dear , Sir
Attached you will find my reviews file, I want to estimate the following equation

CON1 CUR EDU FDI FIN2 INF2 GOE TR UNE Y1

Regards
Attachments
eviewsnew.wf1
(510.78 KiB) Downloaded 213 times

startz
Non-normality and collinearity are NOT problems!
Posts: 3775
Joined: Wed Sep 17, 2008 2:25 pm

Re: Near singular matrix, PMG

Postby startz » Mon Dec 28, 2020 7:08 am

Dependent Variable: CON1
Method: Panel Least Squares
Date: 12/28/20 Time: 06:07
Sample (adjusted): 2000 2018
Periods included: 18
Cross-sections included: 61
Total panel (unbalanced) observations: 755

Variable Coefficient Std. Error t-Statistic Prob.

CUR -8.12E+09 1.86E+09 -4.364800 0.0000
EDU -33963273 27122296 -1.252227 0.2109
FDI -2.22E+08 1.75E+08 -1.269897 0.2045
FIN2 1.26E+08 50515644 2.489889 0.0130
INF2 7.08E+08 1.57E+08 4.512228 0.0000
GOE 5.377901 0.039234 137.0715 0.0000
TR -45312197 35269080 -1.284757 0.1993
UNE -8.21E+08 1.78E+08 -4.618991 0.0000
Y1 1.25E+09 2.73E+08 4.589057 0.0000

Root MSE 2.59E+10 R-squared 0.966630
Mean dependent var 4.67E+10 Adjusted R-squared 0.966272
S.D. dependent var 1.42E+11 S.E. of regression 2.61E+10
Akaike info criterion 50.81722 Sum squared resid 5.07E+23
Schwarz criterion 50.87238 Log likelihood -19174.50
Hannan-Quinn criter. 50.83847 Durbin-Watson stat 0.196020

aawoad
Posts: 8
Joined: Thu Dec 24, 2020 1:49 am

Re: Near singular matrix, PMG

Postby aawoad » Mon Dec 28, 2020 7:25 am

Hello
Thanks, but in my case what is problem with PMG?
Regards

startz
Non-normality and collinearity are NOT problems!
Posts: 3775
Joined: Wed Sep 17, 2008 2:25 pm

Re: Near singular matrix, PMG

Postby startz » Mon Dec 28, 2020 7:32 am

Your workfile did not include the equation you estimated. Either include that equation or post the exact command you used.

aawoad
Posts: 8
Joined: Thu Dec 24, 2020 1:49 am

Re: Near singular matrix, PMG

Postby aawoad » Mon Dec 28, 2020 7:42 am

I selected the same variables that you estimated today, but I selected ARDL/PMG. Could you please help?

startz
Non-normality and collinearity are NOT problems!
Posts: 3775
Joined: Wed Sep 17, 2008 2:25 pm

Re: Near singular matrix, PMG

Postby startz » Mon Dec 28, 2020 7:52 am

Note that you have a lot of NAs. I suspect that once lags are included there isn't any data left.

aawoad
Posts: 8
Joined: Thu Dec 24, 2020 1:49 am

Re: Near singular matrix, PMG

Postby aawoad » Mon Dec 28, 2020 7:58 am

I agree with you, that what I thought, anyway, thanks a lot for help and spport
Regards


Return to “Estimation”

Who is online

Users browsing this forum: No registered users and 15 guests