Censored regression Tobit - Forecast. Error Message

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

KariRingi
Posts: 1
Joined: Fri Nov 17, 2017 12:05 pm

Censored regression Tobit - Forecast. Error Message

Postby KariRingi » Fri Nov 17, 2017 1:45 pm

Hello.

I need to estimate newspaper demand and have therefore made a tobit regression based historical sales. The regression is left censored when it is holiday (since 0 sales) and right censored by a dummy when sales = the delivered quantity for the specific day. However, the estimation of the regression model is fine, but I need to estimate forecast and when I try to I receive the error message "Unable to compute due to missing data" and if I reduce the forecast sample I get the error message "Error in Sample: Range Error."

How can I estimate a forecast of newspaper demand? I added the file in case it is easier to have a look.

I hope someone might know how to do it, any suggestion would be very much appreciated.

Have a great day!

Best regards,
Kari
Attachments
Tobit Regression.WF1
(65.59 KiB) Downloaded 261 times

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

Re: Censored regression Tobit - Forecast. Error Message

Postby EViews Glenn » Mon Nov 20, 2017 10:56 am

For what period do you wish to compute the forecast?

[Edit: I took a closer look at the equation while waiting for a response to the original question and noticed that the censoring is specified by indicator variable. As noted in the manual, in this case, you cannot compute the forecasts of the conditional mean as you do not have censoring information for observations that are not censored. From the manual:
If you specify your censoring by index, you are informing EViews that you do not have information about the censoring for those observations that are not censored. Similarly, if an observation is left censored, you may not have information about the right censoring limit. In these circumstances, you should specify your censoring by index so that EViews will prevent you from computing the conditional mean of the dependent variable and the associated residuals.

Note that you can, of course, compute the forecasted index, which is the forecast of the latent variable.

I hope this clears things up.]


Return to “Estimation”

Who is online

Users browsing this forum: No registered users and 30 guests