04-13-2017 05:43 AM
Hello, I want to analyze simulated data of different sizes in both R and SAS.
I have simulated all the samples in R and where I've done all my analysis, now I want to analyze
all of those samples in SAS. All samples are in a the data frame.
I want to do logistic regression and exact logistic regression,
and I'm interested in the estimated coefficients. (university edition) 1. How do I transfer all samples to SAS smoothly? (So that point 2 gets easy) 2. How do I make a loop so that I can analyze all the data directly?
Note that the sample size goes from 5-1500 with an increment of 5 and for now I
only have one explanatory variable.
Plz help :)
04-13-2017 07:30 AM
1. One easy way would be a csv-file (write.csv, .. in R & proc import in SAS). (And you could simulate your data in SAS as well (?))
2. Depending on your models and data there are different ways too. Check out the "BY" Statement of proc logistic.
04-13-2017 07:48 AM
There are R packages to create datasets directly for example:
Then you simply copy that file to your working area and write code based on that dataset.
AS for your stats question, I don't know the code offhand, but "looping" is done using by group processing.