Friday, December 23, 2005

The curse of choice

As any consumer knows, having many choices can be a curse rather than a blessing. John Kay points out that the choice "between Tweedledum and Tweedledee may not matter much to the chooser but it matters a lot to Tweedledum and Tweedledee". Taking this further, the fact that there are many choices keeps the average quality at a higher level and thus a choice made at random can be expected to be of a high quality than when choice is limited.

Thursday, December 22, 2005

Not all "cluster" diseases are noise

I recently got diagnosed with Lyme disease. So my knowledge of it seems to have grown over the past few days. Here is a trivia that I didn't know about: (cut and pasted from the FAQ on Lyme disease:

1.03 Why is the illness called "Lyme disease?"

"Lyme disease is named after a small coastal town in Connecticut called Lyme, where in 1975, a woman named Polly Murray brought to the attention of Yale researchers an unusual cluster of more than 51 cases of mostly pediatric arthritis. In 1977, Dr. Allen Steere and Yale colleagues identified the new clinical entity and named it "Lyme arthritis." In 1979, the name was changed to "Lyme disease," when Steere and colleague Dr. Steven Malawista discovered additional symptoms linked to the disease: problems of neurologic involvement and severe fatigue.

It wasn't until 1982 that the causative agent of the disease was discovered by Dr. Willy Burgdorfer. Burgdorfer published a paper on the infectious agent of Lyme disease, and earned the right to have his name placed on the Lyme disease spirochete now known as Borrelia burgdorferi. Borrelia burgdorferi (Bb) has since been the official taxonomic name of the Lyme disease spirochete. (Information from Forschner-Vanderhoof K., Everything You Need to Know About Lyme Disease)

Inanity continued

Jonathan Clements writes in his "Getting Going" column that rebalancing should be be done less frequently. It seems that this wrongheaded idea follows entirely from another mistaken idea: that markets can be timed. Momentum strikes again! (cue the soundtrack). Clements even goes so far as to quote William Bernstein (an investment advisor from North Bend, Oregon) who pronounces "that there's significant evidence of momentum in asset class returns". Is this evidence statistically significant? I think not. Clements does make one correct, albeit obvious, point: rebalanceing has tax consequences. duh. Why don't we let the fools be fools and we'll go ahead and invest in equal weighted, frequently rebalanced portfolios (Adi, Mike, Cengiz) or the good old value weighted portfolio (Dean).

Breathtaking Inanity

Judge Jones in his decision to reject the introduction of Intelligent Design into Dover classrooms hits hard: "the breathtaking inanity of the board's decision is evident when considered against the factual backdrop which has now been fully revealed through this trial". Let's hear it for the Judge. Religions can discourage careful scrunity of its details by labeling those who do it from within as sacreligious and from without as intolerant. Coincidence... or Intelligent Design?

Dancing, sex and evolution

Now here's a study we should replicate as an in-class demo! This study shows that people who are thought to dance better are more "evolutionary fit." As in more attractive. Ok, how was it actually operationally measured? By symmetry.

So, how about a dance competition on the first day and an evaluation of how many hook ups were made on the last day of class. If that doesn't get us in to the DP, nothing will!

Thursday, December 15, 2005

Truth as common belief

The Wikipedia has been in the news a lot lately. It tries to generate truth by the law-of-large numbers. If enough people all push the articles a little bit in the right direction, it should drift towards truth.

This has caused lots of hand wringing in the circles that typically protect truth from the masses. Nature recently compared the wikipedia to Britannica. Here is a self-referential truth statement from the wikipedia about it wiki.

Looks like a tie!

Monday, December 12, 2005

The Regression Fallacy and Baseball Salaries

The performance of baseball players is typically measured by accumulated totals of productive batting events. Assuming independence (roughly) among plate appearances implies that recognition is based on estimated success probabilities. Those of us working on the Baseball project know that the variance in those estimates are surprisingly large and typically the same order as the difference between players we consider "good" and those that are ordinary. Thus, the team that buys the star of the hour is almost assuredly overpaying. Is George Steinbrenner listening?

Wednesday, December 07, 2005

Two studies with possibly interesting ethics

Two studies came out this morning that I thought worth mentioning. The first on how strife in a relationship can affect healing times. The second is about IQ and genetics. It appears that if a male has one variant, IGF2R, of a growth factor, that male is 20 points lower on IQ.

So how to make normal science controversial? Its just data folks? The IQ is pretty straight forward--no one likes the idea of say tracking kindergardners based on a genetics test. But it seems like there is potential in the other study that I just can't identify.