How to test c(2)>c(6)

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

cylin
Posts: 3
Joined: Fri Oct 17, 2008 7:54 am

How to test c(2)>c(6)

Postby cylin » Fri Oct 17, 2008 8:06 am

In the Eviews User's Guide, I only find Wald test to test things like C(2)=c(6). How can I test coefficient relationship like c(2)>c(6)?
Thanks.

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

Re: How to test c(2)>c(6)

Postby startz » Fri Oct 17, 2008 8:09 am

cylin wrote:In the Eviews User's Guide, I only find Wald test to test things like C(2)=c(6). How can I test coefficient relationship like c(2)>c(6)?
Thanks.

Do the same test, but you'll have to look up a one-tailed critical value rather than use the p-value reported by EViews. The EViews p-value is for two-tailed tests.

cylin
Posts: 3
Joined: Fri Oct 17, 2008 7:54 am

Re: How to test c(2)>c(6)

Postby cylin » Fri Oct 17, 2008 8:14 am

startz,
Thanks. But how?
For example, I got the following restuls:

Test Statistic Value df Probability

Chi-square 0.129122 1 0.7193


Null Hypothesis Summary:

Normalized Restriction (= 0) Value Std. Err.

C(6) - C(23) 0.027229 0.075777

Restrictions are linear in coefficients.

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

Re: How to test c(2)>c(6)

Postby startz » Fri Oct 17, 2008 8:24 am

cylin wrote:startz,
Thanks. But how?
For example, I got the following restuls:

Test Statistic Value df Probability

Chi-square 0.129122 1 0.7193


Null Hypothesis Summary:

Normalized Restriction (= 0) Value Std. Err.

C(6) - C(23) 0.027229 0.075777

Restrictions are linear in coefficients.


Divide the value by the standard error, from the normalized restriction. This gives the t-stat.

cylin
Posts: 3
Joined: Fri Oct 17, 2008 7:54 am

Re: How to test c(2)>c(6)

Postby cylin » Fri Oct 17, 2008 8:34 am

Starz,
Do you mean:

0.027229/0.075777=0.3593, which is the t-stat?
How can we tell this t-stat is for C(6)>c(23) or C(6)<c(23) ?

ibrahim ben hammadi
Posts: 2
Joined: Fri Apr 02, 2021 2:53 am

test unilatéral

Postby ibrahim ben hammadi » Fri Apr 02, 2021 3:18 am

bonjour,
suite à mes résultats de régression (PIB: variable dépendante, ide; transfert et APD sont des variables indépendantes), j'ai effectué des tests d'hypothèse pour comparer les coefficients individuelles des trois sources de financement ( ide, APD, et transferts). j'ai formé l'hypothèse nulle et alternative par exemple (H0: c(APD)= c(ide) et H1: c(APD)>c(ide)). des tests unilatéraux ont été utilisés.
comment puis-je faire ce genre de test avec Eviews?


Return to “Estimation”

Who is online

Users browsing this forum: Google [Bot] and 29 guests