SAS Support Communities

Thank you Bruno - this is the best workaround solution for now. ... View more

I tried writing to my H: drive, still no luck:   ERROR: Physical file does not exist, H:\HH_Providers.pdf.   The weird thing is that I *can* export datasets to my desktop from EG using the Export wizard. But not when using ods statements to output files from a procedure.   ... View more

Hi Salomon,   Here's the code for both the logistic regression and Poisson regression tested on the Diabetes dataset (available at http://www4.stat.ncsu.edu/~boos/var.select/diabetes.tab.txt).   Either ridge regression (alpha=0), lasso (alpha=1), or elastic net (0 < alpha < 1) may be run by inputing alpha and lambda parameters.   Please note that my code isn't quite complete: in order to decide on an optimal lambda value for the lasso, the standard approach is to use cross-validation to minimize prediciton errors across a range of specified lambda values. To do this, split your analysis dataset into k CV folds (typically k=5 or k=10). Run the program (including the standardization and parameter initialization code) on each of the k training datasets. Then score the validation datasets with coefficients generated from each run. Select the value of lambda with the lowest average prediction error. Another option is the “one-standard-error” rule: select the largest lambda value that is still within 1 stnd error of the lambda value with the minimal prediction error.     ... View more

Ah - in that case maybe a better idea would be to conduct two binomial tests comparing the % Yes for X and Y between the Observed & Expected? ... View more

Thanks Reeza -   Does the "expected" option allow the user to manually enter in the expected null hypothesis values, or does "expected" only calculate the 2x2 table's row and column means through cross-multiplication? ... View more

Thanks RW9, that's an interesting solution.   My code (which performs a lasso/ridge regression) reads the analysis dataset into an array & processes the data in a data _null_ data step to save processing time. Getting it to work was a minor miracle, but there are still a few bugs in my array code & I'm open to suggestions. ... View more

Right, normally I'm happy with the array index starting at 1.   In this case I'm refining a program that iteratively calculates regression coefficients p0, p1, p2, ..., p&numvar, and for convenience's sake I'd like the coefficient array to start at 0 (the intercept). One less thing to keep track of in my code if I don't have to subtract one from the array index when I'm printing out the final array values. ... View more

Thanks Reeza!   I had a feeling there was a simple solution I'd overlooked. ... View more