t+1 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

kerry
Posts: 4
Joined: Mon Oct 05, 2015 10:59 am

t+1 in eviews

Postby kerry » Tue Oct 06, 2015 9:31 am

I'm trying to estimate a simple equation: y(t+1) = a + b*x(t)
Can I put "y(1) c x" into the estimate window? so that (1) will automatically mean t+1?
I've tried both "y(1) c x" and "y c x(-1)" to estimate, I get exactly the same output of regression from these two.
But someone told me that it would be wrong to use y(1) for y(t+1), I have to go back one period and estimate the equation using "y c x(-1)".
Can someone explain to me why? If it is wrong, why am I getting the same output for both method?

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

Re: t+1 in eviews

Postby EViews Gareth » Tue Oct 06, 2015 9:35 am

y(1) is the correct syntax for a one period lead on Y.

kerry
Posts: 4
Joined: Mon Oct 05, 2015 10:59 am

Re: t+1 in eviews

Postby kerry » Tue Oct 06, 2015 9:47 am

y(1) is the correct syntax for a one period lead on Y.
Ok but if i turn ny equation into y = a + b*x(t-1), and estimate using "y c x(-1)",
the intercept and coefficient gotten from this estimate can also be used on the original equation y(t+1)= a + b*x(t) right?
Basically these two equations are the same?

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

Re: t+1 in eviews

Postby startz » Tue Oct 06, 2015 9:50 am

y(1) is the correct syntax for a one period lead on Y.
Ok but if i turn ny equation into y = a + b*x(t-1), and estimate using "y c x(-1)",
the intercept and coefficient gotten from this estimate can also be used on the original equation y(t+1)= a + b*x(t) right?
Basically these two equations are the same?
They are the same equation, but they may differ by one observation in the data used to estimate them.

kerry
Posts: 4
Joined: Mon Oct 05, 2015 10:59 am

Re: t+1 in eviews

Postby kerry » Tue Oct 06, 2015 9:54 am

y(1) is the correct syntax for a one period lead on Y.
Ok but if i turn ny equation into y = a + b*x(t-1), and estimate using "y c x(-1)",
the intercept and coefficient gotten from this estimate can also be used on the original equation y(t+1)= a + b*x(t) right?
Basically these two equations are the same?
They are the same equation, but they may differ by one observation in the data used to estimate them.
so the correct input for my original equation will be "y(1) c x"?
if i remember correctly, few years ago when i first learnt eviews, i was told that eviews does not allow lead on y variables, tat's really strange..


Return to “Estimation”

Who is online

Users browsing this forum: No registered users and 2 guests