Change in model deviance, GLM

[Myers, Nicholas]

Hello.

I write to ask if there is a way within the Generalized Linear Models (GLM) window to get a change in model deviance when adding a predictor. Below I provide the syntax and relevant output (see Step = 6.799) from imposing such a model through the Regression, Binary Logistic window. Note that "delayed" was added in a "sequential" way.

LOGISTIC REGRESSION VARIABLES  cure_di
  /METHOD = ENTER d1 d2 d3 d4 d5  /METHOD = ENTER delayed
  /ORIGIN
  /CRITERIA = PIN(.05) POUT(.10) ITERATE(20) CUT(.5) .


Omnibus Tests of Model Coefficients



Chi-square


df


Sig.


Step 1


Step


6.799


1


.009


Block


6.799


1


.009


Model


46.003


6


.000



Below I provide the relevant syntax for imposing the “full” model (i.e., with “delayed”) through the GLM window. Note that "delayed" is added "simultaneously" with d1-d5.

GENLIN
  cure_cnt OF freq
  WITH delayed d1 d2 d3 d4 d5
 /MODEL
  delayed d1 d2 d3 d4 d5
  INTERCEPT=NO
  DISTRIBUTION=BINOMIAL
  LINK=LOGIT
 /CRITERIA METHOD=NEWTON SCALE=1 COVB=MODEL
  MAXITERATIONS=100 MAXSTEPHALVING=5
  PCONVERGE=1E-006(ABSOLUTE)
  SINGULAR=1E-012
  ANALYSISTYPE=3 CILEVEL=95 LIKELIHOOD=FULL
 /MISSING CLASSMISSING=EXCLUDE
 /PRINT CPS DESCRIPTIVES MODELINFO FIT SUMMARY SOLUTION(EXPONENTIATED).

I believe that I can impose the reduced model (leave out “delayed” in this case) and then take the difference between -2LL from both models to reproduce the Chi-square above. But I wonder if there is a way to get the program to do this (change in model deviance comparing the nested models) for me within the GLM window (or altering the relevant syntax)?

Responses on/off-list wil be appreciated.

Thanks for considering my question.

nick myers

-----------------------------------------------------------------------------------------------------------
Nicholas D. Myers, Ph.D.
University of Miami
School of Education
Department of Educational and Psychological Studies
Research, Measurement and Evaluation program

Merrick Building 311E
P.O. Box 248065
Coral Gables, FL 33124-2040

Tel: [phone]
Fax: [phone]
E-mail: [hidden email]
http://www.education.miami.edu/facultyStaff/Faculty_Bio.asp?ID=143
-----------------------------------------------------------------------------------------------------------

=====================
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