Tuesday, April 27, 2010

This about sums it up

“I'll swallow a lie when I have to; I've swallowed a few big ones lately. But the stat games? That lie? It's what ruined this department. Shining up shit and calling it gold so majors become colonels and mayors become governors. Pretending to do policework while one generation fucking trains the next how not to do the job.”

Thursday, April 22, 2010

Green Ray T-Shirts

GreenRay T-Shirts

Very cool designs, some of which were done by one of my friends (seen modeling the "Matrix Ray").

Wednesday, April 21, 2010

Wednesday, April 14, 2010

Beyond Awesome

Machine learning at its finest:


Sunday, April 4, 2010

Seeing meaning in random market fluctuations

I don't know the exact details, but almost every day on the business headlines I read either that the economy is headed toward disaster because some large corporation fell below its quarterly profit expectation, or that we are headed toward prosperity because a company exceeded it.

For example, from Bloomberg, "Best Buy, the largest U.S. electronics retailer, is among companies seeing demand pick up. The Richfield, Minnesota-based merchant last month reported fourth-quarter profit that exceeded analysts’ estimates as discounts helped boost sales.

Carnival, the biggest cruise-line operator, last month raised its full-year profit forecast as ticket prices rebounded from 2009’s lows amid more bookings."

Are there confidence intervals around these forecasts? Merely exceeding analysts' estimates doesn't seem like news. Furthermore, they provide an explanation--the treatment effect of discounts--that can explain this effect independently of economic recovery.

I guess I don't see how this is news. Does anyone know more about profit forecasting? Are they point estimates or confidence intervals? The same thing with the Dow Jones Index: people seem to interpret random fluctuations as meaningful with absolutely no accounting for margin of error in estimates.