selectionbias

Also approximate Bayesian computation, principal stratification, and the Bayesian Bootstrap.

I used to live in Boston and visited recently for the first time in a while. The red line was so slow between Central Square and Harvard that I thought there must have been some kind of mechanical failure, but when I mentioned it to friends afterwards they said that it had been like that for months.

(Ignore my previous reply I found it myself). To be fair to the authors, it is not their primary specification, that was a linear probability model. The logit model is just a robustness check to make sure the linearity assumption isn't driving the results.

Which paper was this?

Citing a paper needn't be a celebration of it, you can cite a paper to say "these guys find X but..."

Sorry to be pedantic, but although Monte Carlo simulations are based on pseudo-randomness, I still think it is good practice that they have deterministic results (i.e., use a given seed) so that the exact results can be replicated. If the precise numbers can be reproduced then a) it helps me as a reviewer see that everything is kosher with their code and b) it means that if I tweak the code to try something out my results will be fully compatible with theirs.

FYI, the word is 'tenet', a tenant is a person who lives in a particular residence.

Well, at the very least, the argument over hypothesis formation is not specific to economics. The philosopher of science Karl Popper argued that science should aim to empirically falsify hypotheses, and that the way scientists develop those hypotheses, while an interesting psychological question, is irrelevant for the scientific method. He argued that hypothesis formation always contains an element of irrationality and instinct and he quotes Einstein who expressed views to that effect.

I admit I haven't read or even heard of Kate Raeworth, but I am an academic economist. Most modern economic research is empirical, a paper poses a policy-relevant empirical question 'did policy X reduce unemployment' and then empirical evidence is presented (perhaps using data from a randomly controlled trial or using quasi-experimental variation). The statistics are calculated and the assumptions required for the validity of the statistical analysis discussed critically and at great length. Undergraduate econ classes unfortunately leave students with the impression that academic econ is mostly unrealistic theoretical models of behavior, but those models are just handy tools for hypothesis formation, and it is the empirical testing of hypotheses that makes a discipline a science, not the manner by which those hypotheses were formed.

Look at the latest edition of QJE. You will see mostly studies addressing a particular policy question, e.g., 'what was the effect of this policy change on unemployment' which they answer using randomly controlled trials or quasi-experimental methods. Which aspect of this falls apart when you prod it?

My understanding is that medieval Rabbis (notably the Maimonides) explicitly discuss cannabis consumption and its psychological effects. The plant also gets some discussion in the Talmud but in the context of its use in fabric and as candle wick. It is speculated that certain plants mentioned in the Torah refer to cannabis but these strike me as a little tenuous. At the very least the use of cannabis as an intoxicant does not seem to be explicitly mentioned in the Torah. So I wonder, if it was commonly used for that purpose in the ancient middle east, why little to no mention of this? I mean, there is a mountain of detail about all the other minutiae of ancient custom.