Hello, all. I have a two way ANOVA testing the effects of Fungicide and (fungal) Isolate on a Log10EC50 response value (EC50 being the effective chemical concentration that inhibits growth halfway between a minimum and maximum response). I know that my groups of Fungicide*Isolate have very heterogeneous variance (variances range from 0.00073 to 0.87599), and I am trying to fit the model with hetereogeneous variance using the following code.
ANOVA Code
PROC mixed data=IMPORT;
TITLE3 "ANOVA Comparing LogEC50's, group=Isolate*Fungicide";
class Trial Fungicide Plate Isolate;
model LogEC50 = Fungicide|Isolate / ddfm=KR outp=outH;
Random Plate(Trial) / group=Fungicide*Isolate;
run;
I am receiving a memory error. I have a large dataset (over 2,800 observations - the attached PDF shows Class Level Information; there will be about 390 groups that need to be fitted for heterogeneous variance). I have included the fullstimer option to aid in diagnosis.
LOG
ERROR: The SAS System stopped processing this step because of insufficient memory. NOTE: The PROCEDURE MIXED printed pages 4-5. NOTE: PROCEDURE MIXED used (Total process time): real time 0.13 seconds user cpu time 0.06 seconds system cpu time 0.07 seconds memory 5162.28k OS Memory 36272.00k Timestamp 02/25/2016 04:02:39 AM Step Count 31 Switch Count 223 Page Faults 0 Page Reclaims 695 Page Swaps 0 Voluntary Context Switches 898 Involuntary Context Switches 719 Block Input Operations 0 Block Output Operations 40464
ERROR: Integer overflow on computing amount of memory required. A request to allocate the memory cannot be honored.
You could switch to PROC HPMIXED, designed for large-scale problems.
However, I think you want the residual to have different groups, not the random effect.
repeated / group=fungicide*isolate;
In fact, based on the limited information, plate(trial) could be your residual, which would mean that you have an over-parameterized model (since the residual is already there). You might want to try:
random int / sub=trial;
repeated / group=fungicide*isolate;
With all that said, it is usually not good statistical practice to fit so many variances in a model. This usually reduces your power, and can lead to all kinds of estimation problems, even when there is enough memory.
...One more thing.... Quoting from the HPMIXED User's Guide (applies also to MIXED):
"You should exercise caution in defining the GROUP effect, because strange covariance patterns can result with its misuse. Also, the GROUP effect can greatly increase the number of estimated covariance parameters, which can adversely affect the optimization process".
Are you ready for the spotlight? We're accepting content ideas for SAS Innovate 2025 to be held May 6-9 in Orlando, FL. The call is open until September 25. Read more here about why you should contribute and what is in it for you!
Learn the difference between classical and Bayesian statistical approaches and see a few PROC examples to perform Bayesian analysis in this video.
Find more tutorials on the SAS Users YouTube channel.