May 01, 2012

Post Hoc Reasoning

Here's why I'm becoming a Bayesian.

Let's say I collect the winning lottery numbers from the last 10 years. I sped a couple months trying to find some pattern---any pattern---that generates numbers that would have won a few times. It's even possible I might come up with a model that is statistically significant and accounts for some decent percentage of the variance in lottery numbers.

Have I found a way to predict future lottery draws? Will it make me rich? Why not? But it's statistically significant and predicts an acceptable percentage of variance. Why isn't that enough?

In contrast, if I approach the problem from a Bayesian standpoint, the first thing I have to do is defend the notion that such a model can be made: that is my prior probability and I can't run an analysis without it. In other words, the plausibility of the research is part of the research.

There's a downside to this, of course. If people don't believe a hypothesis is likely to be true, the insightful researcher who suspects it is will have a harder time using a Bayesian approach than a traditional approach.

But that's because Bayesian's don't let you get away with post hoc reasoning. And that's a pretty neat trick.

Posted by rbrent at 06:05 AM | Comments (0)

October 12, 2011

The IgNobels!

The IgNobel Prizes for 2011 have been awarded!

If you've never heard of the IgNobels, take some time out to peruse the list of winners, present and past. The prize rewards that strange kind of research that seems silly but is really profound. My favorite this year went to John Perry at Stanford for his 1996 article, "How to Procrastinate and Still Get Things Done". If anyone wants to collaborate on a study examining structured procrastination habits of undergraduate medical students and senior clinicians, let me know: I've been looking for something to do to avoid some more important and pressing tasks.

Posted by rbrent at 07:28 AM | Comments (0)

September 19, 2011

A Scrutiny of the Introduction

Here's a great, short essay by Jon F. Claerbout about how to write a good introduction to a scientific article. From the essay:

"You might be able to produce a good introduction without following my formula but if you have trouble producing one that pleases other people (and you would like to finish it and get on with your life), then I suggest you follow my formula."

Posted by rbrent at 02:59 PM | Comments (0)

Double Header Talk on Thursday 9/22 at 2:00pm

Double header on Thursday, September 22, 2:00pm to 4:00pm in 4448 East Hall (on Church St. between S. University and N. University).

Downs suggests reading this paper:
Strategies for Promoting Healthier Food Choice which has some really cool, unexpected results about the impact of food labeling on health behaviors. From the article:

"In combination, these studies suggest that providing calorie information may have small effects on food choices, but may also produce perverse effects, such as promoting higher calorie consumption among dieters."

Read on!

Posted by rbrent at 12:04 PM | Comments (0)

September 06, 2011

2011-2012 Journal Club is in the planning stages

Here we go! Please feel free to suggest articles.

Posted by rbrent at 02:46 PM | Comments (0)

November 16, 2010

PDF readers

I've been using Skim almost exclusively for a few months now. It's easily the fastest, cleanest, and prettiest pdf reader I've used. I recommend trying it: it's free.

Posted by rbrent at 09:27 AM | Comments (0)