It's EG running locally. ... View more

@bikashten wrote: That's I am doing right now, but I just posted it if there are any alternative better way to do it. It's not fun to do it manually over thousand records. Thanks, Bikash Agreed, but you can't always program your way out of bad data and it's better to fix this at the sources somehow,using a number to identify companies instead of names is a starter, having a cleaned data base, having a verification step as people enter data.    Trying to clean up the mess afterwards is always more work.  ... View more

data test; input id x1 x2 x3 x4; datalines; 1 0 1 3 4 1 1 1 3 4 2 2 4 6 8 2 2 6 6 8 3 3 3 3 3 3 3 3 3 3 ; run; proc sql; create table temp as select id,count(distinct x1) as x1, count(distinct x2) as x2, count(distinct x3) as x3,count(distinct x4) as x4 from test group by id; quit; proc transpose data=temp out=temp1; by id; run; proc sql noprint; select distinct _name_ into : names separated by ',' from temp1 where col1>1; create table want as select id,&names from test; quit; ... View more

Thanks Ksharp for your clarification.  ... View more

   Aska, i think you can use this code for binomial distribution with repeated measures: proc glimmix data=A; class subjectID treatment day; model reactivity=treatment day treatment*day / dist=binomial DDFM=KR; random subjectID; lsmeans treatment day treatment*day/ pdiff lines; /* if there is a significant interaction then you can compare the treatment effects for each day*/; lsmeans treatment*day/sliceby=day pdiff lines; run; Thank you, Bikash ... View more

The problem with doing missing value imputation for PCA is that if the imputation does not take into account the correlation between the variables, then essentially the imputation is going to alter the fitted PCA model. Your code does take this partially into account using FCS LOGISTIC for two variables, but does not take this into account for the correlations between the other variables (like correlation between x2 and x3).   So, I don't really know how to do this using PROC MI. My suggestion is to use PROC PLS where you set the option MISSING=EM (an expectation maximization algorithm is used to replace missing values). To use PROC PLS to get PCA results, the trick is that you have to specify that the x-variables in the PLS model are identical to the y-variables in the PLS model.   But I don't know how well this will work on your data. In addition, you have lots of missing data here, over 20%, and so I would worry that the amount of missing is a problem. ... View more

Hi Ballardw, In this analysis, We don't have any identifiable sub-populations. Simply, we are trying to use the "SRS" method without any strata in proc surveyrselect to create the five non-overlapping groups. It is just a process about sampling procedures to create the non-overlapping groups. Thank you, Bikash ... View more

Hi Ballardw, Thank you so much for replying this thread.   There is also a paper suggesting that scaling the weights to sum to sample size (n) (page 12-http://www.isr.umich.edu/src/smp/asda/J%20Surv%20Stat%20Methodol-2015-Lumley-jssam_smu021.pdf). I am also assuming that if we divide the individuals weight by mean of all weight in order to get new weight scaled to sample size, But I am not sure it is right or not.   Thanks, Bikash ... View more

The CLOGLOG link does not result in a log binomial model. See this note for estimating a relative risk based on a logistic model. You can use an ordinary logistic model (fit in PROC LOGISTIC or SURVEYLOGISTIC), and then use the NLMeans macro as shown there. ... View more

  proc mixed data=               class subject_id position gender stages;               model percent=position gender stages;               repeated position/subject=subject_id;               lsmeans position gender stages run;   is this right? ... View more