Dear NM7 users, According to the examples in the users guide, the IMP step followed the SAEM step with EONLY=1 to evaluate the objective function. In all of my models (for the same data set), I have these two $EST steps: $EST METHOD=SAEM INTER FILE=saem.txt NBURN=3000 NITER=2000 PRINT=5 CTYPE=3 GRD=DDDDDDDDDSSDDD SEED=20091013 ISAMPLE=2 $EST METHOD=IMP EONLY=1 FILE=imp.txt NITER=300 ISAMPLE=3000 PRINT=1
In my base model, the OBJ trend in the IMP step is as follows (truncated here): iteration 296 OBJ= -1796.0722480934633 iteration 297 OBJ= -1802.1845380634943 iteration 298 OBJ= -1795.4946269639131 iteration 299 OBJ= -1795.8286754523187 Elapsed estimation time in seconds: 230513 iteration 300 OBJ= -1791.4880075805286 When I add IOV on one of the parameters (to the base model), I got the following trend, which showed the OBJ became worse: iteration 296 OBJ= 52215.178158178285 iteration 297 OBJ= 51758.149614882743 iteration 298 OBJ= 51850.531967877905 iteration 299 OBJ= 53377.558672094958 Elapsed estimation time in seconds: 314605 iteration 300 OBJ= 53030.267320528052 When I compared the GOF plots and parameter estimates for these two models, they looked pretty much the same. I could not find any obvious differences between them. How would I interpret/compare the OBJ for these 2 models? What could be the reasons for such difference in the OBJ for the 2 models? Thank you in advance for your help, -Thuy
