Can't do simple nonlinear regression

For notifying us of what you believe are bugs or errors in EViews.
Please ensure your copy of EViews is up-to-date before posting.

Moderators: EViews Gareth, EViews Moderator

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

Can't do simple nonlinear regression

Postby startz » Thu Oct 29, 2015 8:57 am

Open the attached workfile

Code: Select all

ls gdpc96=c(1)*exp(c(2)*@trend)


EViews dies
Attachments
gdp and federal funds.wf1
(19.78 KiB) Downloaded 251 times

EViews Glenn
EViews Developer
Posts: 2671
Joined: Wed Oct 15, 2008 9:17 am

Re: Can't do simple nonlinear regression

Postby EViews Glenn » Thu Oct 29, 2015 12:30 pm

One of those bugs that is a combination of things. It's not really a bug in the equation estimation part, but in our handling of an error that occurs because the nonlinear estimation failed.

First your estimation is failing because the starting values for your particular equation are poor (getting starting values when you have exponentials is always a bit tricky as you have to guard against overflow) and the optimizer can't figure out how to start. Normally that doesn't cause problems, but in your case it requires us writing an overflow value into the output table, and we're not handing that operation properly. I've cleaned up the write and a fix will appear in the next patch.

For now, set your starting coefficient values to something safer like 0.1 and you'll be good to go.

Sorry about the inconvenience.


Return to “Bug Reports”

Who is online

Users browsing this forum: No registered users and 21 guests