X12-ARIMA Error with when running regression spec in Eviews

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

kysoon
Posts: 4
Joined: Tue Jul 21, 2009 10:52 pm

X12-ARIMA Error with when running regression spec in Eviews

Postby kysoon » Tue Jul 21, 2009 11:09 pm

Hi,

I am trying to run the following spec file below but constantly get an error message. The odd thing is that the same spec file (with the appropriate series spec added back in) appears to work perfectly in the original X12-ARIMA programme itself, hence this leads me to conclude that there must be something wrong with how I'm passing the spec file to Eviews. Data is monthly data. Help please, I just can't figure it out!

Details are as follows.

ERROR: cny_dur is already in the regression.

Command to run the spec file through Eviews

Code: Select all

freeze(x12out) teu.x12(sfile="C:\Documents and Settings\kysoon\Desktop\X12\teu2.spc", save="d11 d12 d13")

Spec File

Code: Select all

transform{ function=log } outlier{ types=(AO LS TC) lsrun=0 } automdl{savelog=amd} forecast{ maxlead=12 print=none } estimate{print=(roots regcmatrix) savelog=aicc} check{print=all savelog=lbq} regression{ file = "C:\Documents and Settings\kysoon\Desktop\X12\DOS Files\CNY2010.dat" format = "datevalue" variables=( const td) start = 1960.1 user = (cny_bef cny_dur cny_aft) usertype = holiday #aictest=(td easter) savelog=aictest #aictest=(td user) } x11{ savelog=(M1 M7 M8 M9 M10 M11 Q Q2) save=(D11 D12 D13 SP1 SP2) } slidingspans{savelog=percent additivesa=percent} history{estimates=(fcst aic sadj sadjchng trend trendchng)}

EViews Gareth
Fe ddaethom, fe welon, fe amcangyfrifon
Posts: 13604
Joined: Tue Sep 16, 2008 5:38 pm

Re: X12-ARIMA Error with when running regression spec in Eviews

Postby EViews Gareth » Wed Jul 22, 2009 8:11 am

Could you post the CNY2010.dat file?

kysoon
Posts: 4
Joined: Tue Jul 21, 2009 10:52 pm

Re: X12-ARIMA Error with when running regression spec in Eviews

Postby kysoon » Wed Jul 22, 2009 6:20 pm

Attached to this post. Somehow the forum doesn't allow me to upload .dat or .txt files..? So I changed the file extension to .doc, but it is actually just a regular ASCII file.
Attachments
CNY2010.doc
(10.72 KiB) Downloaded 637 times

EViews Gareth
Fe ddaethom, fe welon, fe amcangyfrifon
Posts: 13604
Joined: Tue Sep 16, 2008 5:38 pm

Re: X12-ARIMA Error with when running regression spec in Eviews

Postby EViews Gareth » Thu Jul 23, 2009 8:17 am

I can't seem to replicate your exact error. Could you post your workfile too?

kysoon
Posts: 4
Joined: Tue Jul 21, 2009 10:52 pm

Re: X12-ARIMA Error with when running regression spec in Eviews

Postby kysoon » Mon Jul 27, 2009 12:04 am

Okay, I deleted the rest of my data to leave behind a single series, and am now getting a completely different error.

Code: Select all

ERROR: Unable to open model file for automatic model selection, x12a.mdl. ERROR: Must have user supplied models stored in x12a.mdl.
Attachments
tptdataforum.wf1
(209.71 KiB) Downloaded 476 times

EViews Gareth
Fe ddaethom, fe welon, fe amcangyfrifon
Posts: 13604
Joined: Tue Sep 16, 2008 5:38 pm

Re: X12-ARIMA Error with when running regression spec in Eviews

Postby EViews Gareth » Mon Jul 27, 2009 9:55 am

Great, thanks.

For reasons currently unknown to us, it looks as though it will work if, and only if, you run EViews by launching the EViews6.exe executable directly from within the EViews 6 directory, rather than through a short cut (or through the Start menu in Windows), and then try to perform your X12.

Obviously we're looking into why running EViews 6 from a short cut or the Start menu should make a difference, but until we've worked out why and how to fix it, a solution is to simply navigate to the EViews 6 directory on your computer and then launch EViews 6 from there.

kysoon
Posts: 4
Joined: Tue Jul 21, 2009 10:52 pm

Re: X12-ARIMA Error with when running regression spec in Eviews

Postby kysoon » Mon Jul 27, 2009 6:53 pm

Thanks - it works fine now. Interesting bug.
Perhaps it has something to do with setting the appropriate paths?

EViews Gareth
Fe ddaethom, fe welon, fe amcangyfrifon
Posts: 13604
Joined: Tue Sep 16, 2008 5:38 pm

Re: X12-ARIMA Error with when running regression spec in Eviews

Postby EViews Gareth » Mon Jul 27, 2009 9:22 pm

We weren't setting the working directory properly. We've fixed it in the latest patch.

fmgoto
Posts: 76
Joined: Thu Jan 15, 2009 5:01 am
Location: Brasilia, Brazil

Re: X12-ARIMA Error with when running regression spec in Eviews

Postby fmgoto » Tue Jan 05, 2010 4:12 pm

Gareth, I am also having thesing X12 ARIMA from Eviews. So far, I think X12A has some restrictions on how it is run.

I had some problems running X12A through Excel VBA and it seemed to me that the length of the path is a restriction.

Eviews seems also not to run PICKMDL as well.

Hope Eviews 7 can fix these. Thank You, Fabio

Clopes
Posts: 1
Joined: Thu Sep 20, 2012 5:38 am

Re: X12-ARIMA Error with when running regression spec in Evi

Postby Clopes » Thu Sep 20, 2012 5:44 am

Hi,

I am trying to run the following spec file below but constantly get an error message. The odd thing is that the same spec file (with the appropriate series spec added back in) appears to work perfectly in the original X12-ARIMA programme itself, hence this leads me to conclude that there must be something wrong with how I'm passing the spec file to Eviews. Data is monthly data. Help please, I just can't figure it out!

Details are as follows.

ERROR: cny_dur is already in the regression.

Command to run the spec file through Eviews

Code: Select all

freeze(x12out) teu.x12(sfile="C:\Documents and Settings\kysoon\Desktop\X12\teu2.spc", save="d11 d12 d13")

Spec File

Code: Select all

transform{ function=log } outlier{ types=(AO LS TC) lsrun=0 } automdl{savelog=amd} forecast{ maxlead=12 print=none } estimate{print=(roots regcmatrix) savelog=aicc} check{print=all savelog=lbq} regression{ file = "C:\Documents and Settings\kysoon\Desktop\X12\DOS Files\CNY2010.dat" format = "datevalue" variables=( const td) start = 1960.1 user = (cny_bef cny_dur cny_aft) usertype = holiday #aictest=(td easter) savelog=aictest #aictest=(td user) } x11{ savelog=(M1 M7 M8 M9 M10 M11 Q Q2) save=(D11 D12 D13 SP1 SP2) } slidingspans{savelog=percent additivesa=percent} history{estimates=(fcst aic sadj sadjchng trend trendchng)}

I'm having the same problem but with Eviews 7.


Return to “Estimation”

Who is online

Users browsing this forum: No registered users and 2 guests