Saturday, December 18, 2010

Advice to a Young Statistician

He had grown up with numbers. “My dad
was a truck driver and salesman and a good
amateur athlete. He kept score for the baseball
leagues and the bowling teams, stuff like that,
and because of that I grew up with numbers
around me. He liked doing math – not puzzles,
just numbers.

“And so I grew up always thinking I was
going to be a mathematician or something like
that. I’d get books out of the library – Maths
for the Million, that kind of thing.” He got a
scholarship to Caltech. “I got a real break there.
That was the first year they offered the scholarship,
and but for that I couldn’t have gone.” It
was evidently a remarkable family: all four of
the Efron siblings became academics. “My dad
gave us this pretty clear picture that we weren’t
suited for heavy work.”..

Bayesian methods are fine, but if you get too far into Bayesian
methods you quit thinking about inference because it all becomes automatic

Statisticians work at two basic levels. They can develop statistical methods, like linear models, or they can prove things about inference properties. The first is the one that makes you wildly popular with
people who use statistics for their work; I like to work at the second level.

In some ways I think that scientists have misled themselves into thinking that if you collect enormous amounts of data you are bound
to get the right answer. You are not bound to get the right answer unless you are enormously smart.
You can narrow down your questions; but enormous sets of data often consist of enormous numbers of small sets of data, none
of which by themselves are enough to solve the thing you are interested in, and they fit together in some complicated way.
-Interview with Brad Efron (Stanford statistics professor)

No comments:

Post a Comment

Note: Only a member of this blog may post a comment.