May be you could find an approximation to the probit function that was faster. There is this stackexchange thread that might be useful, in particular the 3rd answer has an approximation and a link to a paper that discusses more options.  Or perhaps a hybrid approach - if a lot of your probability values are small, it might be possible to find a very good and fast approximation for values less than some threshold (1E-4 say), and then use the probit function for everything else. ... View more

2. Is there some way to call EXPANDGRID with a varying number of arguments?   You could possibly build the the command that you want in a string and then use 'call execute'.  But I note that what you want is essentially counting in binary, so you could create an IML module for this as follows : proc iml; start binmat(p); a = 0:(2##p - 1); b = cshape(putn(a, cat("binary",char(p),".")), 2##p, p, 1); return( 2#(num(b) - 0.5) ); finish; print (binmat(5)) [format=2.0]; quit;   ... View more

I don't think I can offer much help.  You can see the source code for the function ExpMatrix from the IML library.  Look for the library catalog called SASHELP.IML, and then you can double-click on ExpMatrix to view the source, it is possible that you are using a different algorithm in VBA that would account for the differences. ... View more

Please would you show us the syntax you are using both in IML and in EXCEL? ... View more

I can see at least 4 places where you are using syntax like x = x // y  inside a loop.  This 'growing' of a matrix by repeated concatenation is usually very inefficient as a new version of x needs be created for each iteration.  It is much better to declare x at its final size and then fill it up as the loop iterates.  ... View more

Many thanks for posting your code, it looks useful. You might want to add options symbolgen; to your code which will write the values of the macro variables inside the submit block to the log, making it easier to understand what it is going on. ... View more

Well you could always 'submit' the include statement from within IML like this: filename pcode "&mypath\procprint_code.sas"; proc iml; file pcode; put 'proc print data=sashelp.class;'; put 'run;'; closefile pcode; submit; %include pcode; endsubmit; quit ;   ... View more

Executefile() will not do what you want as it is designed for including SAS/IML statements from a file.  There are probably two options to do what you want - take a look at the submit/endsubmit statements which you can use to run general SAS statements from within an IML program.  If this does not have the flexibility you need, then create a file from within IML that later is %included after IML has finished running.  For example: filename gcode "&mypath\graphics_code.sas"; proc iml; file gcode; nplot=3; put 'proc sgplot data=sgplot_data;'; do i=1 to nplot; s = 'series x=x' + strip(char(i)) + ' y=y' + strip(char(i)) + ';'; put s; end; put 'run;'; closefile gcode; quit ; %include gcode; ... View more

You will need about 3GB to store a 20000x20000 matrix and this is probably more than your default memory allocation for SAS.  To see run: proc options option=memsize value; run; There is a command line option -MEMSIZE that you use when starting the SAS application to increase the allocation.  Alternatively, describe why you need to create such a large matrix, and what you want to do with it, then perhaps someone can suggest a way of doing it that needs less memory. ... View more

Just to clarify the call you should be making to analyze you own data should be something like: r = mlr("dataset", {"x1" "x2" "x3"}, "y"); ... View more

If you have a function module which returns a value, then you must assign the result to a variable as follows: r = mlr("sashelp.class", {"Height" "Age"}, "Weight"); print (r$1); In your code lines 2 to 5 are not doing any useful and should be removed.   ... View more

You are not running any IML code here, all you are doing is defining a code module called 'mlr'.  I suggest moving the USE and the two READ statements to the end of the program, after the finish statement, and then after that call the code module using run mlr(x, y); If you do this, it may work or at least give you an error that can be diagnosed. ... View more

@PaigeMiller & @SteveDenham , Instead of dividing the variables into 2 groups like this, why not specify all of the variables both as predictors and responses? When X=Y, PLS on complete data reduces to PCA in the sense that you get the same scores and loadings from each matrix and they match PC scores and loadings. So I have been using the missing=EM syntax, with incomplete data tables and a high number of iterations, to obtain EM-PCA single imputations. You don't even need to make copies of your variables as PROC PLS does not complain when the same variable names appear on the left and right sides of the model statement. ... View more

The syntax "matrix ** 4" is taking the 4th power of the matrix - so the whole matrix multiplied by itself 4 times.  This might not be what you intend and could potentially lead to an overflow. ... View more