The DO Loop
Statistical programming in SAS with an emphasis on SAS/IML programs
I often blog about the usefulness of vectorization in the SAS/IML language. A one-sentence summary of vectorization is "execute a small number of statements that each analyze a lot of data." In general, for matrix languages (SAS/IML, MATLAB, R, ...) vectorization is more efficient than the alternative, which is to
Last week I received a message from SAS Technical Support saying that a customer's IML program was running slowly. Could I look at it to see whether it could be improved? What I discovered is a good reminder about the importance of vectorizing user-defined modules. The program in this blog
SAS procedures usually handle missing values automatically. Univariate procedures such as PROC MEANS automatically delete missing values when computing basic descriptive statistics. Many multivariate procedures such as PROC REG delete an entire observation if any variable in the analysis has a missing value. This is called listwise deletion or using
The SAS DATA step supports multidimensional arrays. However, matrices in SAS/IML are like mathematical matrices: they are always two dimensional. In simulation studies you might need to generate and store thousands of matrices for a later statistical analysis of their properties. How can you accomplish that unless you can create
In SAS, the order of variables in a data set is usually unimportant. However, occasionally SAS programmers need to reorder the variables in order to make a special graph or to simplify a computation. Reordering variables in the DATA step is slightly tricky. There are Knowledge Base articles about how
A SAS/IML programmer asked a question on a discussion forum, which I paraphrase below: I've written a SAS/IML function that takes several arguments. Some of the arguments have default values. When the module is called, I want to compute some quantity, but I only want to compute it for the