I am new to utilizing SAS 9.4 full time for analyses. I am working on my dissertation data and I know that the best way to analyze my data is a generalized linear mixed model. No one else in my lab utilizes SAS, they prefer SPSS, which does not to as good a job with very complex stats in my opinion. My data is a secondary analysis of a year long dataset on women who were undergoing an intervention to regain menstrual function. My Y (resume) is did they or didn't they resume menses during the study. I am interested in whether body composition (of which I have 12 variables) changed the odds of menstrual recovery and am not interested in the intervention effect (rnd) for this analysis. My main concern is fitting the model so that the subject-specific random intercept is assumed and uneven repeate structure of the data (and number of measures per subject) is considered in the analysis. I am not sure the solution I have come up with following extended reviews of the GLIMMIX literature reflects the analysis I am interested in. Proc Glimmix data=resumption; Class period id rnd; Model resume = wt bmi perbf fm fmi lmi lpct tpct tlpctr apct gpct agpctr/ link=logit dist=binomial or solution; Random int/subject=id type=ar residual; Run; My data is set up like: Options nocenter pageno=1; Data resumption; Input period $ id rnd wt bmi perbf fm fmi lmi lpct tpct tlpctr apct gpct agpctr resume; Cards; SBI1 1 3 45.90 16.07 12.30 5.99 2.10 14.37 12.58 9.81 0.78 6.40 9.57 0.67 0 I5 1 3 50.90 17.82 14.90 7.74 2.71 15.81 16.20 12.57 0.78 10.37 21.34 0.49 0 SBI1 14 2 52.16 19.88 17.60 9.29 3.54 16.66 20.74 14.42 0.69 13.01 25.17 0.52 0 I5 14 2 53.90 20.54 18.60 9.97 3.80 16.72 22.38 14.70 0.66 13.96 27.56 0.51 0 I9 14 2 53.00 20.20 17.70 9.48 3.61 16.93 21.74 13.78 0.63 12.73 25.93 0.49 0 SBI1 126 2 50.23 17.87 24.30 12.22 4.35 12.71 31.10 19.19 0.62 . . . 0 SBI1 132 3 56.45 20.09 23.76 13.25 4.71 14.21 29.03 20.95 0.72 . . . 0 I5 132 3 56.41 20.07 . . . . . . . . . . 0 I9 132 3 56.50 20.21 26.25 14.74 5.24 13.88 30.40 24.86 0.82 . . . 0 SBI1 155 2 40.95 16.81 10.10 4.05 1.66 14.04 12.50 6.90 0.55 5.00 17.00 0.29 0 I5 155 2 40.45 16.60 12.20 4.89 2.01 13.67 14.70 9.40 0.64 7.20 20.00 0.36 0 I9 155 2 41.20 16.91 12.10 4.90 2.01 13.82 15.20 9.00 0.59 7.30 20.80 0.35 0 I21 155 2 40.20 16.50 11.00 4.41 1.81 13.83 13.40 8.20 0.61 5.40 17.60 0.31 0 I33 155 2 39.00 16.01 7.50 2.92 1.20 13.98 8.60 4.80 0.56 3.90 11.00 0.35 0 I49 155 2 42.10 17.28 9.80 4.12 1.69 14.81 11.40 7.40 0.65 5.80 16.40 0.35 0 SBI1 170 2 56.20 23.77 32.40 18.02 7.62 14.94 30.70 35.40 1.15 45.10 40.20 1.12 0 SBI1 175 3 54.35 20.33 19.60 10.65 3.98 15.50 24.00 16.80 0.70 18.20 31.30 0.58 0 I5 175 3 55.85 20.89 20.10 11.43 4.28 16.21 24.40 17.40 0.71 18.60 31.30 0.59 0 I9 175 3 56.60 21.17 20.40 11.55 4.32 16.06 24.70 17.70 0.72 19.70 32.60 0.60 0 I21 175 3 56.85 21.27 21.40 12.11 4.53 15.86 25.90 19.00 0.73 21.80 33.10 0.66 1 SBI1 217 3 51.30 18.64 23.00 11.66 4.24 13.45 28.90 18.10 0.63 18.20 34.90 0.52 0 I5 217 3 52.70 19.15 25.20 13.16 4.78 13.48 30.10 21.40 0.71 22.60 37.70 0.60 0 I9 217 3 54.25 19.71 26.80 14.17 5.15 13.31 31.00 23.40 0.75 26.70 39.60 0.67 0 I21 217 3 54.55 19.82 29.10 16.34 5.94 13.76 35.10 25.10 0.72 28.40 44.00 0.65 0 I33 217 3 54.55 19.82 28.40 15.57 5.66 13.55 34.10 24.80 0.73 27.00 43.10 0.63 0 I49 217 3 55.20 20.06 27.30 14.70 5.34 13.46 32.80 23.40 0.71 24.60 41.40 0.59 1 The time periods I have body composition variables are not evenly distributed (screening, and intervention weeks 5, 9, 21, 33, and 49). Duration of time to resumption is not of interest in my analysis. Participants that resumed menstrual function did so at various time points, and many did not resume. Not all of the participants made it through the study to intervention week 49. Not all women have the apct, gpct, and agpctr variables due to a change in the machine used to evaluate body composition. The apct and gpct have shown to be significantly different at the time of resumption compared to non resumers when analyzed with a Hotelling's T2 test, therefore I want to keep these variables in the analysis. I understand that most of my variables are correlated; however, only 7 of the variables are correlated above a rho = 0.95 and none are at 0.99. Thanks for any support!
... View more