EViews.com

Hi,

For my thesis, I am researching the influence of terrorism on financial markets. Thereby, I study 68 different countries, for which I have written a simple program to generate the same regression for all of these countries. But for some reason, I get the error "Near singular matrix" only for country60(which is Switzerland), although I don't see any difference in the data or the dummies used. In attachement you can find the workfile and program, just for country60. Can someone please tell me what the problem could be? If it would help, I can also upload the data for all 68 countries. Thank you very much,

Yasmin

The regressors are perfectly collinear for these data. You'll get a singularity if you run least squares using the specification. To confirm, I also made a group with the regressors, performed a covariance analysis on it and saved the SSCP matrix after removing means. That matrix is singular.

Thanks Glenn!

And what can I do to solve this? I'm not a real expert in EViews, you see...

Yasmin

It isn't an EViews issue. You're trying to do something that is mathematically impossible.

well, isn't the solution "exclude variables that are multi-collinear and re-estimate with different variables"?


Aysen

It depends on your purpose. But if you're interested in getting the right coefficient values, then no. Multicollinearity indicates there isn't enough information to identify separate effects.

Dear Eviews expert,

Thank you so much for your answer on near singular matrix. I am currently using Eviews 5.1 but will install 7 soon. Can you also help me with this warning? When I run hausman test to determine fixed or random effects estimator, I get this warning.

What should I do to filter the problem? Since Hausman test probability (0.68) is larger than critical value (0.05), it is appropriate estimate with random effect (as I was told, if test value < p value of 0.05, select fixed effect in panel options).

However, I get this warning. Should I just go ahead and estimate with random effect in panel options based on the p value?

Thank you so much in advance for your help.
Aysen


Correlated Random Effects - Hausman Test
Equation: EQ01
Test cross-section random effects

Test Summary Chi-Sq. Statistic Chi-Sq. d.f. Prob.

Cross-section random 10.058132 13 0.6892

** Warning: estimated cross-section random effects variance is zero.

Cross-section random effects test comparisons:

Variable Fixed Random Var(Diff.) Prob.

D((EFF)) -0.010736 -0.012434 0.000001 0.0543
D(LEV) 0.002818 0.001137 0.000015 0.6675
D(CRIS) -0.313009 -0.311921 0.000075 0.9001
D(LIQ) -0.015282 -0.006224 0.000029 0.0904
D((TA)) -0.000000 -0.000000 0.000000 0.4669
GDPERC -0.000004 -0.000004 0.000000 0.6398
HHI index 0.307426 0.300954 0.004346 0.9218
D(MG) 0.033089 0.034236 0.000002 0.4050
D((RSIZE)) 0.000000 0.000000 0.000000 0.8341
MCAPGDP -0.008102 -0.008532 0.000001 0.6317
D((NONIXEA)) -0.177128 -0.144414 0.000545 0.1611
CR4 0.040030 0.049236 0.001080 0.7794
D((MS)) -0.089909 -0.067109 0.002804 0.6667

Dear Eviews expert,

Thank you so much for your answer on near singular matrix. I am currently using Eviews 5.1 but will install 7 soon. Can you also help me with this warning? When I run hausman test to determine fixed or random effects estimator, I get this warning.

What should I do to filter the problem? Since Hausman test probability (0.68) is larger than critical value (0.05), it is appropriate estimate with random effect (as I was told, if test value < p value of 0.05, select fixed effect in panel options).

However, I get this warning. Should I just go ahead and estimate with random effect in panel options based on the p value?

Thank you so much in advance for your help.
Aysen


Correlated Random Effects - Hausman Test
Equation: EQ01
Test cross-section random effects

Test Summary Chi-Sq. Statistic Chi-Sq. d.f. Prob.

Cross-section random 10.058132 13 0.6892

** Warning: estimated cross-section random effects variance is zero.

Cross-section random effects test comparisons:

Variable Fixed Random Var(Diff.) Prob.

D((EFF)) -0.010736 -0.012434 0.000001 0.0543
D(LEV) 0.002818 0.001137 0.000015 0.6675
D(CRIS) -0.313009 -0.311921 0.000075 0.9001
D(LIQ) -0.015282 -0.006224 0.000029 0.0904
D((TA)) -0.000000 -0.000000 0.000000 0.4669
GDPERC -0.000004 -0.000004 0.000000 0.6398
HHI index 0.307426 0.300954 0.004346 0.9218
D(MG) 0.033089 0.034236 0.000002 0.4050
D((RSIZE)) 0.000000 0.000000 0.000000 0.8341
MCAPGDP -0.008102 -0.008532 0.000001 0.6317
D((NONIXEA)) -0.177128 -0.144414 0.000545 0.1611
CR4 0.040030 0.049236 0.001080 0.7794
D((MS)) -0.089909 -0.067109 0.002804 0.6667

any comment would be appreciated

Dear eviews experts,

Thanks fo the usefull information provided previously...I still have some questions.

I am running an egarch model to determinate the volatility of an interest rate and I would like to include specific events dummies (I am using daily data and eviews 7). However, I am having a multicolinearity problem.

I am not very used to eviews and would like to learn more. I have checked the correlogram matrix and vif in stata and they dont show a problem with my event variables(see results below). However when using eviews and regressing my dependent variables on the events, a problem of multicollinearity appears to exist (message "near singular matrix"). I also tried dropping the constant or one of the dummies, but the problem continues.

I also have calculated the correlation matrix of my event variables in eviews and a problem is pointed out as I get NA as result.

I enclose my workfile.

Would anyone help me understand:

1) Is the multicollinearity problem also related related to the fact that my event dummies are almost always zero, and only 1 for a few observations (7, 5 & 168 obs out of a total of 3425) ?

2) why I do I get NA when calculating the correlation matrix?

Thanks a lot for your help!

Here are results on VIF correlogram matrix (from stata):

. vif

Variable | VIF 1/VIF
-------------+----------------------
ev_0510 | 1.03 0.971557
ev_3 | 1.03 0.971657
ev_1 | 1.00 0.999894
-------------+----------------------
Mean VIF | 1.02

. correlate ev_1 ev_3 ev_0510
(obs=3425)

| ev_1 ev_3 ev_0510
-------------+---------------------------
ev_1 | 1.0000
ev_3 | -0.0017 1.0000
ev_0510 | -0.0103 0.1684 1.0000

At your current sample, all observations are zero.

true! I guess I have to get more used to eview, thanks for answering!