[Limdep Nlogit List] FW: Homogenity Test page R11-17
Laurence
Laurence.Lester at flinders.edu.au
Thu Mar 22 13:03:48 EST 2007
Can anyone please tell me why, after running the procedure for the
homogeneity test (R11-17) several times successfully with one data set
further attempt to run (exactly the same syntax for the proc without an
change to namelists) give the errors --immediately following the execute;
proc line:
--> EXECUTE; PROC = Samemodl(REGRESS, Y, X, i) $
Error 32: Model command. Start vals. unreadable or wrong # given.
Error 449: DO... Syntax error. Expected DO..;label;control $
Oddly, yesterday I ran the proc several times and after a few it stopped
working (error as above) even with the same data set (I used the airline
data with a very simple model).
The syntax (straight out of the manual) is-- CALC;G=MAX(GROUP);LOGLU=0$
DOFOR;100;GRP=1,G$ INCLUDE;NEW;GROUP=GRP$ MODEL;LHS=Y;RHS=X$
CALC;LOGLU=LOGLU+LOGL$ ENDDO;100$ SAMPLE;ALL$ MODEL;LHS=Y;RHS=X$
CALC;LOGLR=LOGL;LIST;LRTEST=2*(LOGLU-LOGLR);DF=(G-1)*KREG;PROB=1-CHI(LRTEST,
DF)$
I have not been able to run it without an error for my own data. I have
tried a very simple regression (depended is LM_L1, just 2 continuous
explanatory variables, with 5 periods (cwave) as the index, or two groups
LSIA) without any success. Data are as describe below:
Descriptive Statistics
All results based on nonmissing observations.
============================================================================
===
Variable Mean Std.Dev. Minimum Maximum
Cases
============================================================================
===
----------------------------------------------------------------------------
---
All observations in current sample
----------------------------------------------------------------------------
---
LM_L1 .288695130 .453795257 -1.00039838 1.00048066
15540
AGE 35.3088160 8.78871188 15.0000000 81.0000000
15540
URATE 7.00838481 3.41639010 .400000000 32.1000000
15540
CWAVE 2.69884170 1.34500334 1.00000000 5.00000000
15540
LSIA 1.27953668 .448786012 1.00000000 2.00000000
15540
I have tried all sorts of things, deleting loglu, loglr etc, putting in the
index manually, etc but it won't work. If I reload the airline data it will
work again for for several runs the gives the errors.
Regards
Laurence
Laurence Lester
National Institute of Labour Studies
Flinders University, SA
Ph: (08) 8201 2002
Email: Laurence.lester at flinders.edu.au
Mail: GPO Box 2100, Adelaide, SA 5001
More information about the Limdep
mailing list