Forecasting not possible - missing data

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

maca
Posts: 4
Joined: Sat Nov 24, 2012 8:29 am

Forecasting not possible - missing data

Postby maca » Sat Nov 24, 2012 8:44 am

For our studies, we are supposed to model a stock regression. We came up with a model, which we assume is quite good and we were confident about our output.
The equation is
Stock C Rent3 GDP3 Empl3 Takeup4 Vac3 Int3

Our task is to forecast a period from 2012 to 2016. We only have full data of GDP, Employment and Takeup (Forecasts 2012-2016). Thus, we are missing the forecasts for Vacancy, Rent and Interest. Our actual data is from 1982 until 2011.
When computing the forecast in Eviews, an error message popped up saying "unable to compute due to missing data"
What can we do? Help, please! Attached you'll find our work and the excel sheet covering the dataset.
REMF_project_Option_B(2).xls
Excel Sheet with dataset
(60 KiB) Downloaded 301 times
LondonOffices.wf1
Eviews File for London Offices
(46.74 KiB) Downloaded 246 times

maca
Posts: 4
Joined: Sat Nov 24, 2012 8:29 am

Re: Forecasting not possible - missing data

Postby maca » Sat Nov 24, 2012 8:45 am

THANK YOU,
MACA

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

Re: Forecasting not possible - missing data

Postby EViews Gareth » Sat Nov 24, 2012 10:26 am

You need data for those other variables.

maca
Posts: 4
Joined: Sat Nov 24, 2012 8:29 am

Re: Forecasting not possible - missing data

Postby maca » Sat Nov 24, 2012 4:53 pm

what do you mean? we can only use the data given....

startz
Non-normality and collinearity are NOT problems!
Posts: 3798
Joined: Wed Sep 17, 2008 2:25 pm

Re: Forecasting not possible - missing data

Postby startz » Sat Nov 24, 2012 5:03 pm

To make a forecast, it's necessary to multiply estimated coefficients by the right hand side variables. So either you need to confine yourself to variables that will be known in the future, or you need to find a way to forecast the right hand side variables before making the forecast for the left hand side variable.

Maybe that's what the assignment is trying to teach you?

maca
Posts: 4
Joined: Sat Nov 24, 2012 8:29 am

Re: Forecasting not possible - missing data

Postby maca » Mon Nov 26, 2012 9:57 am

got it now.... thanks!


Return to “Estimation”

Who is online

Users browsing this forum: No registered users and 2 guests