A friend recently asked me why I am writing a book. My answer? Some people are born to write a book and some have books thrust upon them. Mine was thrust upon me, although it is more accurate to say that I thrust it upon myself. My book, Statistical Programming
Missing values are a fact of life. Many statistical analyses, such as regression, exclude observations that contain missing values prior to forming matrix equations that are used in the analysis. This post shows how to find rows of a data matrix that contain missing values and how to remove those
Peter Flom reminded readers of his blog that you should always end a SAS procedure with a RUN statement. This is good rule. However, PROC IML is an exception to the rule. In PROC IML, the RUN statement is used to execute a built-in subroutine or a user-defined module. You
A frequently performed task in data analysis is identifying all the observations in a data set that satisfy certain conditions. For example, you might want to identify all of the female patients in your study or to identify all patients whose systolic blood pressure is greater than 140 mm Hg.
"How do I apply a format to a vector of values in IML? In the DATA step, I can just call the PUTN function.” This question came from a SAS customer that I met recently at a conference. My reply? Use the PUTN function, but send it a vector of
The R You Ready blog posed an interesting problem. Essentially, you have a vector that contains n(n+1)/2 elements, and you want to pack those elements into the upper left triangular portion of a matrix. For example, if your data are proc iml; /** vector v is given: ncol(v) = n(n+1)/2 for
When programmers begin learning a new computer language, the first program they write is often one that prints the text “Hello, World!” Successfully writing a Hello World program assures the programmer that the software is successfully installed and that all necessary features are working: parsers, compilers, linkers, and so on.
I just returned home from Vancouver, British Columbia, where I attended the 2010 Joint Statistical Meetings (JSM). I heard that more than 5,300 statisticians attended this year, including about 40 or so from SAS. I stayed busy. I gave a presentation on techniques for visualizing time series, gave a two-hour