You would have to write some macro code to pick out which by groups have VIF>=5 and then print only those. You can capture the VIF output into a SAS data set using ODS OUTPUT statements. To get some idea of which variables might be correlated, I would use Principal Components Analysis on the independent variables. ... View more

People can have multiple years of birth? Are you asking to find where a row has all three years 2000 2001 2002 or where a row has values for all three variables YOB_1 YOB_2 YOB_3? Message was edited by: Paige ... View more

I don't think you can do this in "one quick statement" — at least not according to my definition of the word "quick". You could probably do this in a Macro do-loop within PROC SQL. Another possibility is to do this in a data step. In a data step, you can merge the two data sets together and then do lots of addition of corresponding columns of variables using ARRAYs. Not what I would call "quick", but it works. Message was edited by: Paige ... View more

The only time statements like "var x1 has 20% var x2 has 12%" have meaning is when the variables are not correlated with one another. As soon as you have correlations, statements of the sort "var x1 has 20% var x2 has 12%" are meaningless. I suggest you adopt a different mindset. You can still determine which combinations of variables are important in the prediction equation; you cannot determine which individual variable is causing the impact via statistics alone, nor can you quantify the percent impact of an individual variable. One technique that allows you to determine the combinations of variables that are important in the predicting a response(s) is Partial Least Squares regression. ... View more

I look forward to your blog entries. I have used IML for probably 20 years, but I'm sure there's a lot more I can learn ... View more

You can do this, but you will have to fix the other independent variables to a certain value. When you plot dependent variable against time, you are plotting the main effect only. If there are interactions, this plot will not indicate interactions. ... View more

PROC STEPWISE? My search of support.sas.com seems to indicate there is no such procedure. B or Beta? What does that mean? ... View more

Not everything is identical in SQL implemented for Oracle and SQL implemented in SAS. There are differences! ... View more

data have reject; do id=1 to 3; if id=2 then output have; else output reject; end; run; ... View more

There are plenty of examples as support.sas.com. http://support.sas.com/documentation/cdl/en/proc/61895/HTML/default/viewer.htm#/documentation/cdl/en/proc/61895/HTML/default/a000146767.htm ... View more

I have no idea what your data step creating data set HAVE is doing. it looks like you are using a loop over three ids to create a one record where id=2 and var='' and the other records have var='A'. Totally meaningless example to me, and I can't understand what the rejected records are that you are discussing. Clarifications would be helpful. ... View more

HelenSAS: are you talking about PROC GPLOT or PROC PLOT? ... View more

plot y*x=symbolvar; ... View more

You want to use the DIRECTORY and DETAILS options of PROC CONTENTS; along with DATA= _ALL_ ... View more

So let's take this idea of not pre-quoting macro variables a step further. Suppose I want to set up a macro that does the quoting and creating of a string for later use, so I don't have to edit Cynthia's %file1 macro each time I want to change the SQL. In other words, I want to put a %MEND after Cynthia's %do i = 1 %to &numstr; %if &i lt &numstr %then %let inlist = &inlist.&&word&i., ; %else %if &i eq &numstr %then %let inlist = &inlist.&&word&i; %end; This could result in a general macro that I can use prior to any PROC SQL where such a list is needed. My solution, so far, has been to make &INLIST a %GLOBAL variable inside this macro, and then in the PROC SQL outside of the macro, I insert the %GLOBAL variable &INLIST into the SQL WHERE clause. Is there a better way to handle this? ... View more