Log likelihood

[T.J.C. Beebee]

I'm trying to discover how to do something that's probably very
simple...how to obtain log-likelihood estimates of regression models. The
linear standard regression output of SPSS doesn't seem to provide this. Can
anyone advise, please?
Thanks
Trevor Beebee

Prof. Trevor J.C. Beebee
Department of Biochemistry
School of Life Sciences
University of Sussex
Falmer
Brighton BN1 9QG
UK
phone: [phone] ext 2690
fax: [phone]
email: [hidden email]

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[Bruce Weaver]

You could use MIXED to run your linear regression model--it uses maximum likelihood estimation (or REML) rather than ordinary least squares, and the -2LL is included in the output.  Here's an example using the 1991 General Social Survey data file that comes with SPSS.


REGRESSION
  /STATISTICS COEFF OUTS CI R ANOVA
  /DEPENDENT prestg80
  /METHOD=TEST (age) (educ)  .

MIXED prestg80  WITH age educ
  /METHOD=ML
  /FIXED = age educ | SSTYPE(3)
  /print = solution testcov
.

T.J.C. Beebee wrote

I'm trying to discover how to do something that's probably very
simple...how to obtain log-likelihood estimates of regression models. The
linear standard regression output of SPSS doesn't seem to provide this. Can
anyone advise, please?
Thanks
Trevor Beebee

Prof. Trevor J.C. Beebee
Department of Biochemistry
School of Life Sciences
University of Sussex
Falmer
Brighton BN1 9QG
UK
phone: [phone] ext 2690
fax: [phone]
email: [e-mail]

=====================
To manage your subscription to SPSSX-L, send a message to
[e-mail] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD