About SAS_Rob

SAS_Rob · ‎12-07-2021

Without seeing your code I would say that this is likely due to a large fraction of missing information (FMI). You should expect an increase in the variance (and thus a reduction in significance), specifically the between imputation variance, when the FMI is high. This section of the documentation will be helpful in that regard. SAS Help Center: Multiple Imputation Efficiency The other possible cause is that you have a bad imputation model (in the Proc MI step) or there is non-convergence in the MI models. SAS Help Center: Checking Convergence in MCMC If you can post your code and LOG (including the MI, modeling and MIANALYZE steps) then there might be something more concrete I can suggest.

SAS_Rob · ‎12-02-2021

Why are you adding Pills303 to the BY statement? To have the same variable on both the BY statement and in the MODEL statement is going to cause singularity issues since for each value of Pills303 it will fit a different model (and the values are constant). Likewise, the ESTIMATE statement does not make sense since there is only a single coefficient for Pills303 and it looks like you are trying to compare two separate levels..

SAS_Rob · ‎11-27-2021

Yes, they can be different even when there is no weight variable.

SAS_Rob · ‎11-23-2021

To answer your question regarding the harm, you will be biasing your results by ignoring any of the design effects. To what extent they are bias can be somewhat quantified by calculating the design effect. https://go.documentation.sas.com/doc/en/pgmsascdc/9.4_3.4/statug/statug_surveylogistic_details40.htm

SAS_Rob · ‎11-23-2021

This has to do with the manner in which the SE themselves are calculated. SURVEYLOGISTIC uses (by default) a Taylor Series approximation which readily accomodates weights, strata and clusters. SURVEYLOGISTIC is built for complex survey data so when you don't have any of the design effects adjusted for, there is no reason to use it. The documentation has more details on the variance estimation. https://go.documentation.sas.com/doc/en/pgmsascdc/9.4_3.4/statug/statug_surveylogistic_details27.htm#statug.surveylogistic.taylordetails LOGISTIC on the other hand uses maximum likelihood for the variance estimation. https://go.documentation.sas.com/doc/en/pgmsascdc/9.4_3.4/statug/statug_logistic_details05.htm

SAS_Rob · ‎11-17-2021

Because the theory behind the combining of estimates from multiply imputed data requires both a point estimate and a standard error (which you don't have in this case), you cannot get a pooled -2 LL measure using MIANALYZE. I have not seen a discussion anywhere in the literature as to if or how this might otherwise be done, but maybe someone else in the Community has some suggestions.

SAS_Rob · ‎11-17-2021

Very often you will see things such as salary reported on a different scale for this reason--ill conditioned data due to large units. All you end up doing is changing the units by rescaling so you would just report the results in the new units.

SAS_Rob · ‎11-17-2021

Combining ESTIMATE statement results can be done following this example. data test; do a=1 to 3; do rep=1 to 10; x=ranuni(123)*10; if ranuni(234)>.55 then y=.; else y=.46+.88*a+x+rannor(123); output; end; end; proc mi data=test out=outmi; var a x y; run; ods trace on; proc genmod data=outmi; by _imputation_; class a; model y=a x; estimate 'A1_v_A2' a 1 -1 0;*do not use spaces in the labels; estimate 'A1_v_A3' a 1 0 -1; ods output estimates=estimate_ds; run; proc sort data=estimate_ds; by label _imputation_; run; proc mianalyze data=estimate_ds; by label; modeleffects LBetaEstimate; stderr stderr; run;

SAS_Rob · ‎11-17-2021

Could you also post the full output from: proc means data=ch3ex min max mean std; var salary Time Cits; run; I suspect that, given the magnitude of the estimates, the data itself might be ill-conditioned, but this would confirm that.

SAS_Rob · ‎11-16-2021

You have a bunch of extra variables on the CLASS statement in MIANALYZE. You should only be using variables involved in the model from GENMOD. PROC MIANALYZE parms(classvar=level)=gm_fcs; class civil depression alcohol ed mobility unsafe money worksite ; MODELEFFECTS INTERCEPT pills303 agyw ed civil alcohol drugs depression mobility worksite pimp violrec homeless money unsafe estigmasw estigmahiv astigmasw astigmahiv disclose; run; Also, since you are using a REPEATED statement in GENMOD you need to make sure that you are feeding the empirical estimates into MIANALYZE and not the MLEs. In other words, your ODS OUTPUT statement is incorrect. Use this one instead. ods output GEEEmpPest=gm_fcs;

SAS_Rob · ‎11-16-2021

Post your code and we can see if there is anything unusual about it.

SAS_Rob · ‎11-16-2021

Can you run the following and post the output and LOG? PROC reg data=ch3ex; model salary = Time Cits/collin; run;

SAS_Rob · ‎10-25-2021

Here is a simple example of how to normalize the weights so that they sum to the sample size--you would then use the new weights in the WEIGHT statement in SURVEYLOGISTIC /*Create a Sample Dataset to work with*/ data test; seed=2534565; do i=1 to 1000; weight=i/7; x1=ranuni(seed); logit=-2 + 2*x1; p=exp(-logit)/(1+exp(-logit)); if ranuni(seed)>p then y=1; else y=0; output; end; run; /*Calculate the sum of the weights*/ proc means data=test noprint sum; var weight; output out=weight_sum sum=sum_wt; run; /*Calculate the normalized weight for each observation*/ data normalize;set test; if _n_=1 then set weight_sum; normwt=_FREQ_*(weight/sum_wt); run; proc surveylogistic data=normalize; weight normwt; title 'SURVEYLOGISTIC with the normalized weights that were created'; model y=x1; run;

SAS_Rob · ‎10-15-2021

I suggest you open a track with SAS Technical Support. Be sure to supply your exact operating system, SAS Site Number and Precise release number. If you can share the program and data that is causing the issue then that will also help them figure out what is going on. https://www.sas.com/en_us/contact/technical-support.html

SAS_Rob · ‎10-15-2021

There is a defect in the the Intel Math Kernel Library (MKL), which SAS/IML uses to accelerate certain computations in a multi-threading situation (like in matrix multiplication). Unfortunately there is not much we can do to avoid the problem directly since Intel has not fixed the problem. There is a workaround for this ERROR message : Put the following statements before and after the IML step: options NOTHREADS; proc iml; . . . quit; options THREADS; This prevents IML from creating multiple threads when multiplying large matrices and invoking the Windows MKL functions that are defective. There will be a loss of performance, but you should avoid the ERROR message.

Online Status	Offline
Date Last Visited	‎11-26-2025 09:18 AM

Re: Comparing pooled medians for multiple groups

Re: Running R in SAS

Re: Comparing pooled medians for multiple groups

Re: PROC MI MINIMUM and MAXIMUM can I get MI to show the MIN|MAX assoc...

Re: PROC PHREG

Re: PROC PHREG

Re: PROC PHREG

Re: PROC PHREG

Re: "Invalid reference value" error for PROC CAUSALMED

Re: "Invalid reference value" error for PROC CAUSALMED

Re: Comparing pooled medians for multiple groups

Re: how to use the macro after the imputation?

Re: Proc Surveylogistic and Contrast line

Re: FIsher's Exact test zero row/column

Re: Pooled descriptive statistics after multiple imputation

Re: Comparing pooled medians for multiple groups

Re: Running R in SAS

Re: PROC MI MINIMUM and MAXIMUM can I get MI to show the MIN|MAX assoc...

Re: How to use proc mianalyze after proc mi in a mixed variables case?

Re: How to use proc mianalyze after proc mi in a mixed variables case?

Re: proc mianalyze

Re: PROC MIANALYZE freezes SAS

Re: Proc surveylogistic and proc logistic

Re: Proc surveylogistic and proc logistic

Re: Proc surveylogistic and proc logistic

Re: Is it possible to generate a pooled -2 Log Likelihood from proc mi...

Re: PROC CAUSALMED Warning messages

Re: PROC MIANALYZE freezes SAS

Re: PROC CAUSALMED Warning messages

Re: PROC MIANALYZE freezes SAS

Re: PROC MIANALYZE freezes SAS

Re: PROC CAUSALMED Warning messages

Re: Normalizing Survey Weights in PROC SURVEYLOGISTIC?

Re: SAS 9.4 IML 15.1 crashes

Re: SAS 9.4 IML 15.1 crashes