Waylaid Dialectic

November 4, 2012

A brief after lunch nerd break…

Filed under: Random Musings — terence @ 12:21 pm

…I need to get back to my spreadsheets but:

Ezra Klein is interesting.

Brad Delong is funny.

Natalia Cecire is thought-provoking but wrong.

Some of the comments attacking her are grim but commenter polymath offers an excellent, lucid defence of statistical modelling.

Personally the main problem I have with Cecire’s post is that she conflates horse race analysis with statistical work more generally. It is as if she thinks the only quant work ever done on elections is about predicting outcomes.

And sure, that’s what Nate Silver does. But qual types gaze at running horses too. Newspapers have plenty of pundits offering non-number related prognosticating on the election. And quants do lots of stuff other than watching the track.

Contra Cecire, statistical work can, for example, provide us with information on the roll of race in the US election.

Quantitative analysis has its limitations. Sure. But think Cecire’s post would, I think, have been better if she’d spent more time thinking about what they really were and less time calling people puerile.


Create a free website or blog at WordPress.com.

%d bloggers like this: