Reading

What I have been reading:

"Google Co-Founder Has Genetic Code Linked to Parkinson’s" (New York Times)

Studies show that his likelihood of contracting Parkinson’s disease in his lifetime may be 20 percent to 80 percent, Mr. Brin said.

Talk about useless statistics.  A confidence interval that is utterly useless.

"How Wall Street Lied to Its Computers" (New York Times)

So where were the quants?

Risk manager must be the most miserable job ever.  When traders were raking in the millions, quants didn't get the credit (or the pay), according to Taleb, etc.  Now when the market is imploding, they get the blame?

"Competing Tax Plans: two perspectives" (Freakonomics blog)

Three ways to plot the distribution of tax cuts across income brackets.  I don't see why the first, and simplest, chart has a problem.  The two revisions use bar charts with varying-width bars which give excessive focus on the number of people, in one case, and the base income, in the other case.  It is not easy to compare areas of a tall, thin bar and a narrow, flat bar.  The income group labels also present a problem of "loss aversion": why not lose the precision? or just report the percentiles?

Comments

[Ken]

The problem that Wall Street didn't consider was that their investments were changing the whole economy. Debt was rising massively, until of course the first default happened. Then the money flows reduced and there wasn't the same available to pay debt.

They simply maximised for the local system and then found out that the global mattered.

[nate]

If you're interested in the better statistic:

"One recent study found that a person who inherits the mutation has a 28% chance of developing Parkinson’s by the age of 59, 51% by the age of 69 and 74% by the age of 79."

http://spittoon.23andme.com/2008/09/18/google-co-founder-blogs-about-23andme-data-parkinsons-risk/

[Jon Peltier]

Re the Freakonomics charts. Don't look at them as bar charts, look at them as step charts. Not as good as a smooth continuous curve, but it shows the distribution in other ways without giving equal weight (equal bar thickness) to grossly unequal sample/bin sizes.

[Ali]

"I don't see why the first, and simplest, chart has a problem."

It doesn't really, as far as the data it's trying to show. The problem is trying to convey the extra information about how many people are affected in each tax bracket. The reason is how many people actually get this wrong. Charlie Gibson famously failed to appreciate this in *two* elections -- debates in 2004 (Kerry v Bush) and 2008 (Obama v Clinton). Somehow, the Democratic talking point of top 1% or whatever never makes it through (according to poll, I think). Personally, I think most voters aren't really thinking about taxes as much as other issues. Democrats worry otherwise. I think this is why the second chart is getting a fair bit of play in the liberal blogosphere.