Math Forum / Math Software / SAS / July 2009

Am I the only one who finds the the SAS 9.2 documentation harder to
navigate than it used to be in previous versions?
And what's the story with PROC SETINIT?
It still exists, is still referred to in notes like http://support.sas.com/kb/31/290.html,

hi  ... I don't think that anyone has posted a link to this ...
Tuesday, July 28, 2009
IBM buys SPSS: Good for analytics, but what about customers?
http://blogs.sas.com/jimdavis/

I notice the Noonan press release uses the phrase Predictive Analytics (his capitalization) no fewer than three times.
Seems to be a clear shot across the bow.  I'm sure this isn't the last we'll hear about this.
Mike Rhoads
RhoadsM1@Westat.com

I working on a RUG paper
and have a test program on this page
http://www.sascommunity.org/wiki/Evaluating_Logical_Expressions
that may show that logical evaluation proceeds from right to left

Ron
My recent favorite around the office is the politically incorrect phrase "waterboarding the data".
Nat
-----Original Message-----

I want to plot response rates across time for a number of studies. However I
would like to know if it is also possible to have the symbols indicate
whether a specific variable is either 0 or 1. I would like to do this
without labeling, i.e. varying symbol size. Is this possible?

I have a list of URLs for which I would want to extract the titles for
the pages linked to those URLs. Can some one help me in this regard.
example: www.gmail.com - GMail - would the page title.
Can anyone help me?

data have;
  input ID  year X method;
  cards;
1   2000  0.9  1

I am trying to select observations based on the maximum value within a by
group.  I have data such as:
ID  year  X    method
1   2000  0.9  1

I usually use google group http://groups.google.com/group/comp.soft-
sys.sas/topics to search old SAS-L post, because it's much powerful than
SAS-L itself. But today, I tried to search a simple key word "DDE",
it only returned 4 entries. It doesn't make sense at all since DDE was

I'm sorry to keep harping on this issue, but despite having done quite
a bit of research I still have not come up with an answer.
I recently posted a strange (to me) phenomenon that occurred when
running essentially a repeated measures logistic regression with a

I'm hoping someone can explain to me if it is ok or not to transform a
dependent variable, conduct an analysis, then back transform the
estimate and the endpoints of a CI in a linear model?
I. I have heard this is not OK with ANOVA or linear regression, say LN

I have been asked to check out a company called Space-Time Research
and compare it to SAS and Enterprise Guide.  I have never heard of
them.  Does anyone have any information or links to info about them
that would help, e.g comparison, history of company, etc.  I have the

In PROC SORT, can we choose which duplicate record to keep by a conditional
statement?
For example, my data looks like this
Date      Y

I don't know how I can create a macro as a product of 3 previous
macros.
I have:
%LET BandRec=5;