Mixed model output

[Fabienne Crettaz]

I have a problem with SPSS output of a mixed model with random intercept and slope. Precisely I don't understand why SPSS gives me a unique estimate of variance for the two random effects instead of two variances (one for intercept and the other for slope). My design includes a response variable (y), a between-factor (type, 2 types of therapy, new or tau), a within-factor (time, 3 measures of response at month 1, 4 and 12) and a covariable (baseline, measure of response y before intervention). Finally, time and baseline are continuous varaibles. My syntax is: MIXED y BY type WITH time baseline /CRITERIA=CIN(95) MXITER(100) MXSTEP(10) SCORING(1) SINGULAR(0.000000000001) HCONVERGE(0, ABSOLUTE) LCONVERGE(0, ABSOLUTE) PCONVERGE(0.000001, ABSOLUTE) /FIXED=time baseline type | SSTYPE(3) /METHOD=REML /RANDOM=INTERCEPT time | SUBJECT(Code) COVTYPE(ID). Here is the part of the oupout posing problems: Estimations des paramètres de covariance(a) Paramètre Estimation Erreur standard Résidu 32,130446 3,008434 Constante + time [subject = sujet] Variance ,035415 ,033983 Thank you for any help

[Maguin, Eugene]

I think your covtype selection may be wrong. In your situation I’d use UN. Look at page 1115 in the 23 syntax reference.

Gene Maguin

 

 

 

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Fabienne Crettaz
Sent: Wednesday, July 15, 2015 5:07 AM
To: [hidden email]
Subject: Mixed model output

 

I have a problem with SPSS output of a mixed model with random intercept and slope. Precisely I don't understand why SPSS gives me a unique estimate of variance for the two random effects instead of two variances (one for intercept and the other for slope). My design includes a response variable (y), a between-factor (type, 2 types of therapy, new or tau), a within-factor (time, 3 measures of response at month 1, 4 and 12) and a covariable (baseline, measure of response y before intervention). Finally, time and baseline are continuous varaibles. My syntax is: MIXED y BY type WITH time baseline /CRITERIA=CIN(95) MXITER(100) MXSTEP(10) SCORING(1) SINGULAR(0.000000000001) HCONVERGE(0, ABSOLUTE) LCONVERGE(0, ABSOLUTE) PCONVERGE(0.000001, ABSOLUTE) /FIXED=time baseline type | SSTYPE(3) /METHOD=REML /RANDOM=INTERCEPT time | SUBJECT(Code) COVTYPE(ID). Here is the part of the oupout posing problems: Estimations des paramètres de covariance(a) Paramètre Estimation Erreur standard Résidu 32,130446 3,008434 Constante + time [subject = sujet] Variance ,035415 ,033983 Thank you for any help

---

View this message in context: Mixed model output
Sent from the SPSSX Discussion mailing list archive at Nabble.com.
===================== 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

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