Breusch-Pagan LM Test for Random Effects

[gxg 10]

hello. i have a panel data. i have to run breusch pagan lm test to test the random effects from the ols but i cant see the test in eviews 7. pls help me

[cocktailzz]

hey Gareth,

i'm Delta, i'm at the last year of university, and i've to finished my research.

i've read your post, and it's related to my problem right now. actually my research is consist of one dependent variable and two independent variables.
by purposive sampling method i've got 8 samples, and the time period of each sample is 5 years. 8 x 5 = 40 observation.

and now, i have to test whether i'ts better to use common effect or random effect (LM test).
actually i've tried to run the program u have attached. but i have no idea how to switch the variable with my research variable. means like i couldn't adjust the program.

could you help me to fix it please, Gareth?

here is my excel file, i need your help to import the data , estimate the signification of the variables using eviews in common and random effects model, after that compare them by using LM test (need your help to adjust the program code), just until this step, after that i'll do it by my self.

i beg you Gareth, i really need your help... :(

[EViews Gareth]

What have you tried so far?

[nadiasorayaputmeu]

Helloo ..

I have use this program and i got the probability result is 0.05066 so i used Common effect , am i wrong ? :D

[EViews Glenn]

I just wanted to interrupt the discussion to point out that since this thread first started, we have released EViews 9 which features built-in tools for computing the Breusch-Pagan LM test, along with the Pesaran scaled LM, Pesaran CD, and the Baltagi, Feng, and Kao bias corrected scaled LM tests.

Sorry for the interruption. Carry on.

[adnma]

Glenn I am using EVIEWS 8...and installed the add in for Breusch Pagan LM test.

Howver, when I run the test, I am gettinga message as:

Procedure can only be run from equations estimated by list

Why is this happening??

Kindly advice me in this regard.....

[adnma]

Hi!

An update to my previus post....In one of the posts, I read that if we just list the variables and (as Y X1 X2)then run it might work....I tried but still am getting the same message. I am using EVIEWS 8

Please help me....