Monday, September 30, 2013

empirical economics

Two studies one of my empirical economist friends should do:

1) Identify the selection bias in motorcycle fatality/crash statistics by comparing to scooters, which are objectively just as dangerous but attract a very different kind of rider. (All kinds of supplemental sources of identification and controls might also be useful here - state-by-state laws on what kinds of bikes and scooters are highway legal, state-by-state laws on related things such as helmets or lane-splitting, changes in the above kinds of laws, comparing specific types of crashes, such as taking turns too fast or being hit by other cars, which are unlikely to have anything to do with whatever added difficulty there is from having to shift manually or having a higher center of gravity, specific sizes / styles of bikes / scooters, introduction of new types of scooters into the market, comparing specific types of usage such as urban commuting vs. recreational, age and experience of riders, etc. An actual empiricist would probably come up with cleaner ideas.)

2) Identify the impact on job market outcomes from jet lag by using the random location of the AEA conference each year. Number of follow-up interviews, ultimate placement, etc.

Wednesday, September 25, 2013


Freedom, Jonathan Franzen - Couldn't put it down. This guy wrote at least a dozen complex well-developed characters, any of whom would've been the best written character in most novels. It's 570 pages but not nearly long enough.

Longitude: The True Story of a Lone Genius Who Solved the Greatest Scientific Problem of His Time, by Dana Sobel - After visiting the Greenwich Observatory, which is essentially a museum of longitude, my friend Henry told me about this book which goes into more detail. Pretty fascinating. Amazing how many thousands of people used to die (ship wrecks, getting lost at sea, etc), not even that long ago, due to such a simple thing that we take completely for granted now.

Introvert Power: Why Your Inner Life Is Your Hidden Strength by Laurie Helgoe - Painfully naval-gazey, written entirely in therapist-speak. I'm just about as introverted as you can get (not antisocial but 100% introverted according to every Myers-Briggs test I've ever taken) so I can empathize with every single sentence, but that also means there wasn't a single new sentence. And I'm long past the angsty teen self-exploration phase of my life where it would've been entertaining as part of a self-definition clarification thing.

Friday, September 20, 2013

sociology-economics phrasebook

This is perhaps the best thing I've ever seen. [HT MR]


need                                                 want
is correlated with                            is correlated with
determines                                       is correlated with
causes                                             is correlated with
crosstabs                                         non-parametric regression
empirical work                                crosstabs
structural analysis                            OLS regression
sophisticated structural analysis      logit model
endogenous                                     endogenous
exogenous                                        endogenous
model (I)                                           explanation
model (II)                                         diagram involving circles and arrows
exploitation                                      contract
discrimination                                  wage differential

Ok ok I'll stop before I quote the whole thing. TOO good. (And entirely true! I kept reading sociology papers with "model" in the title/abstract hoping they would have something useful to say about my models, since, you know, it seemed like a good idea not to be completely out of the loop on what sociology says about my very sociological economics dissertation, but the best I ever found was the diagram with circles and arrows...)

Thursday, September 19, 2013

re-SMBC, part 4

Well this one was easy.


(Although, if I were to get pedantic, I'd say the assumptions of microeconomics are rarely exactly true but operate as analogies of very common real-world situations, and the resulting models do really well at explaining those situations. Macro is the same way but the resulting models don't do so well. Perhaps because the assumptions are way too far from the truth, I dunno. I'm not gonna pretend to be a macroeconomist. I barely remember what IS-LM stands for...)

Another note: behavioral economics does better at explaining things partly because it's kind of an increase in the degrees of freedom, but I very much like Matthew Rabin's point on this matter: lambda=0 is the implicit assumption of classical economics, but lambda=2 is a much better assumption and doesn't increase the degrees of freedom at all.

Thursday, September 12, 2013

the ppt panic

I confess... Anytime powerpoint crashes or fails to load a plugin or can't find an attached file during a presentation, I experience uncontrollable schadenfreude. Especially when the speaker is openly disdainful or skeptical of TeX. (Which in economics is basically an if and only if condition for using powerpoint in the first place.)

This has happened at least three times in the last month :)

(Hey, at least my tech snobbery hasn't gotten to the point where I begrudge others' use of things like non-linux or non-emacs or their fear of non-Matlab... I just feel bad that they're stuck torturing themselves. The TeX thing is more irritating because they're likely to try to inflict their .doc's on me :)

Wednesday, September 11, 2013


This is what I did while clicking refresh on the AEA registration website for an hour this morning. Re-SMBC part 3:


(Context and more good jokes.)