Anyone who claims to know all of SAS is clinically insane  – Ernest Hemingway (not intended to be a factual attribution)
Ernest Hemingway
Okay, I admit it, Hemingway didn’t really say that, but he would have, except for being dead and all.

As usual, this year I didn’t have time to do everything I wanted at the Western Users of SAS Software conference. As usual, I spent the majority of my time when I wasn’t presenting in the Statistics and Analytics section. Also as usual, I didn’t get to attend any presentations by David Pasta. He presents every year and I have READ several of his papers on line afterward and they are really good . For some reason, his papers are always at the exact time that I am presenting or when I have a meeting with someone. Yet another reason to go back next year and try again.

As usual, there were so many presentations I wanted to attend that I did not get outside for days. As one programmer said to another,

“I went outside once. The graphics kind of sucked.”

Buildings in SF
Unlike most years, nothing came up in the papers I attended that I could immediately use – unless you count Maura Stokes mentioning that ODS GRAPHICS ON will be the default in SAS 9.3 – which had me go up to the hotel room and add an * to several slides for my presentations the next two days where a program began with ODS GRAPHICS ON to have a footnote at the bottom saying no longer necessary when you get SAS 9.3.

Even though I didn’t find anything I could go back and use on Monday, which is unusual, that doesn’t mean at all that the conference wasn’t really worth attending. Very often, I will learn something and three months or six months later, I’ll have a need for PROC X. PROC GLMSELECT , also mentioned in the presentation by Maura Stokes (I did go to other presentations, really) is one I haven’t used before. It is used (surprise surprise) to help you select the best model. According to the documentation, “The procedure offers extensive capabilities for customizing the selection with a wide variety of selection and stopping criteria, from traditional and computationally efficient significance-level-based criteria to more computationally intensive validation-based criteria. The procedure also provides graphical summaries of the selection search. ” It was way cooler when you see the results than it sounds like from that description.

PROC VARCLUS was mentioned by David Pasta in the Q & A after one of the papers. That is another procedure I haven’t used, but that I want to explore further. According to the documentation “The VARCLUS procedure divides a set of numeric variables into either disjoint or hierarchical clusters. Associated with each cluster is a linear combination of the variables in the cluster, which may be either the first principal component or the centroid component.”

Something I can see coming in handy sooner rather than later is the %GetTweet macro to analyze (what else?) tweets from twitter, discussed in a paper by Satish Garla who is a student at Oklahoma State University.

Really interesting to me was the Java object talk by Diane Shealy. I did not even know there was a Java object in SAS. Apparently, the Javaobj has been hiding from me for a long time because once I started looking into it, I found an interesting paper by Richard DeVenezia about using JavaObj in the DATA step that was written back in 2004.  We are likely to be doing a lot with Java in our company over the next year so if there is an easy way to connect it to SAS, I am intensely interested. Diane is also a student.

What is it with these bright, motivated students? Didn’t they read that article by Joel Stein in Time magazine this week saying they were the lazy generation?

When I was their age I was drinking beer at frat parties, not presenting at conferences.

So, there you have it, my homework from WUSS

GLMSELECT
VARCLUS
%GetTweet
Javaobj

Even cooler, Martha McLean  on twitter had mentioned #readingweek. When I asked what that was she said it was something they had at the university she attended and now she applies it to her work. Every now and then she sets aside two or three days and does all the reading of work-related blogs, articles and books she has been meaning to read.

What an awesome idea. I hereby declare November 1-7 my reading week. Don’t try to stop me.

Comments

One Response to “My Reading Week Schedule, thanks to WUSS”

  1. Martha on October 15th, 2011 9:47 am

    Well done!! ;)

    Most Canadian universities had reading weeks although I daresay many folks used it to head to the sunny shores of the US!

    My list and number of bookmarks, favourties, read-it-laters continues to grow – clearly I’m not the only one! I’ll likely be joining the early November #readingweek. Mine will see me book a few days off work (as part of
    my learning plan) and soak it all up.

    Your list is impressive – look forward to hearing how your week goes!!

    Enjoy
    @mjmclean

Leave a Reply