Grocer Function
NAME
arlm - Lagrange multiplier autocorrelation test
CALLING SEQUENCE
[resulbp]=arlm(resul1,p,np)
PARAMETERS
Input
-
resul1 = results tlist from a first stage estimation
-
p = # of lag of residuals in the second stage estimation
-
np = 'noprint' if the user does not want to print the results
Output
-
resulbp = results tlist with:
- resulbp('meth') = 'archtest'
- resulbp('resul1st') = resul1
- resulbp('f') = fstat
- resulbp('p') = p
- resulbp('df') = df
- resulbp('f_pvalue')=f_pvalue
DESCRIPTION
Computes a Lagrange multiplier autocorrelation test of order p (see Godfrey, L. G. (1978). Testing for higher order serial correlation in regression equations when the regressors include lagged dependent variables. Econometrica, 46, 1303-1313). Results are stored in a tlist and displayed on screen.
EXAMPLE
rbp=arlm(rols,4)
This example is taken from hendryericsson. It provides the Lagrange multiplier autocorrelation test of order 4 for Hendry and Ericsson preferred equation, whose estimation result has been saved in tlist rols.
AUTHOR
Eric Dubois 2002