## Near singular matrix - D-F

For econometric discussions not necessarily related to EViews.

Moderators: EViews Gareth, EViews Moderator

Facundo
Posts: 3
Joined: Tue Nov 13, 2018 7:46 am

### Near singular matrix - D-F

Hi,

I simulate an AR(1) time series with coefficient "1.2". Then when I run the Dickey Fuller test, I get "Near singular matrix" message. I suppose that it has to be with Ho (coefficient=0) and H1 (coefficient < 1), but I can't explain it.

startz
Non-normality and collinearity are NOT problems!
Posts: 3443
Joined: Wed Sep 17, 2008 2:25 pm

### Re: Near singular matrix - D-F

You might want to post your workfile and exact commands.

Facundo
Posts: 3
Joined: Tue Nov 13, 2018 7:46 am

### Re: Near singular matrix - D-F

I found that the problem is the maximum lag length to consider when performing automatic lag length selection. If I set it on '0', there is no problem; if it is higher, I get the message.
Here is the code. Thanks

Code: Select all

`'Workfilenew workfile ejercicios q 1951:01 2000:04'White noiseseries e=nrnd'TS-simulationsmpl 1951:01 1951:01series y1=0smpl 1951:02 1951:02smpl 1951:02 @lastseries y1= 2*y1(-1) + esmpl @all'Unit Root Test - ADFy1.uroot(adf, const, trend, maxlag=0) 'maxlag=0, OKy1.uroot(adf, const, trend, maxlag=1) 'maxlag=0, near singular matrix`

startz
Non-normality and collinearity are NOT problems!
Posts: 3443
Joined: Wed Sep 17, 2008 2:25 pm

### Re: Near singular matrix - D-F

I believe you're getting numerical problems. Note that the final values of y1 are around 10^60

Facundo
Posts: 3
Joined: Tue Nov 13, 2018 7:46 am

### Re: Near singular matrix - D-F

Thanks, it worked with less observations.

### Who is online

Users browsing this forum: Google [Bot] and 6 guests