I've not encountered any problems, either. Given that you can open the file, could it be that SAS is exporting it as .xlsx, but the others can only read .xls? ... View more

I agree. Shall I formally report this to SAS, or is posting here good enough? ... View more

PG, Here's an interesting twist--if you have only 1 BY group, the code works.  Columns are named with an underscore before the year, e.g., _2008. Mark ... View more

You want to calculate a phi coefficient as a measure of association for binary data.  If you run the usual Pearson correlation in Proc Corr on binary data, the measure you get will be the phi coefficient, as they are equivalent. ... View more

Proc surveyfreq will calculate the design effect for proportions by using the DEFF option in the TABLES statement. Mark ... View more

I think you will need to do this in more than one step: 1) draw your first stage sample of n=15, using pps and without replacement; 2) merge your sample back with the original data to get your second-stage data included; 3) draw your second stage sample of n=30 from the merged sample, using the first stage information as a stratum, not a cluster. Surveyselect will draw 30 samples from each of your 15 first-stage samples if you treat them as strata. Regards, Mark ... View more

This will calculate the overall mean of avgreturn and the result will appear in your output window. You can also use proc univariate and proc summary to do the same (proc summary produces an output dataset only). proc means data=datasetname mean; var avgreturn; run; Regards, Mark ... View more

The simplest method to model seasonal patterns is to use dummy variables to represent the seasons. With such a short time series, a weekly periodicity is about the longest effect you can detect (i.e., you would not be able to see a monthly effect with only 4 months). A regresion model with the seasonal dummy variables as your inputs will serve you, but if there is any residual autocorrelation remaining, your MSE may seriously underestimate the variance of the error terms, so your p-values could be way off. I'd use proc autoreg with the dummy variables and ask for the Durbin-Watson statistic to check for autocorrelation. Proc autoreg can handle autoregressive errors, if present. Mark ... View more

I don't think it can be done. SAS uses the corrected sums of squares and crossproducts of the residuals to obtain a partial correlation, and you need the original data for this. http://support.sas.com/documentation/cdl/en/procstat/63104/HTML/default/viewer.htm#procstat_corr_sect017.htm ... View more