10 Final LMM: Diagnostics

10.1 Model assumptions

\(\Huge u_{0i} \sim N(0, \sigma_P^2)\)

\(\Huge e_{ij} \sim N(0, \sigma_e^2)\)

10.2 SPSS inputs

Figure: Display parameter predictions for random effects

Figure: Save predicted values and residuals

10.3 SPSS: Script

MIXED SBP BY Sex Group WITH SBP0 Age BMI Time 
  /CRITERIA = DFMETHOD(SATTERTHWAITE) CIN(95) MXITER(100) MXSTEP(10) 
    SCORING(1) SINGULAR(0.000000000001) HCONVERGE(0, ABSOLUTE) 
    LCONVERGE(0, ABSOLUTE) PCONVERGE(0.000001, ABSOLUTE) 
  /FIXED = Sex Group SBP0 Age BMI Time | SSTYPE(3) 
  /METHOD = REML 
  /PRINT = G R SOLUTION 
  /RANDOM = INTERCEPT | SUBJECT(ID) COVTYPE(VC) SOLUTION
  /EMMEANS = TABLES(OVERALL) 
  /EMMEANS = TABLES(Sex) 
  /EMMEANS = TABLES(Group) 
  /SAVE=FIXPRED SEFIXP PRED SEPRED RESID.  

10.4 Model outputs

10.4.1 Histogram of residuals

10.4.2 Q-Q plot of residuals

10.4.3 Plot of residuals against fitted values

10.4.4 Q-Q plot of random effects