Hacker News Reader

Comment by wjnc | Hacker News Reader

It's an educational course in quantitative economical methods. Fitting real world data is messy and would probably distract. There obviously is overlap with metrics but as an undergrad course I'd separate this too. They do have ample links to scientific papers that do use real world data. There's a pages long list of references [1]. Do check them out if you're into economical science.

[1] https://lectures.quantecon.org/jl/zreferences.html

View on HN

yeah, doing real empirical stuff is messy. but that's exactly why people should learn on real data, and not in a sandbox.

I let my kids practice with hammers, nails and wood (tools / supplies) before I introduce building a piece of furniture (the educational end-goal). These models are the tools of the trade.

I agree with the sentiment that if work is messy, teaching should have messy as well. But not when you're starting out with new tools.

+1 for the kids tools. To extend, I let my kids practice hammering first, then sawing, then some other skill. Learning to work with messy data can wait until you are used to the new tools.

It is literally impossible to run these models on data without first understanding these methods deeply, because causal inference for these observational data IS extremely difficult. If you do not understand your model deeply, or if you do not understand your data deeply, you are likely producing garbage. This course relates to the first point.

There's a lot of structural econometric papers that do exactly what you ask, but you need graduate level statistics and a deep understanding of discrete choice, identification and simulation methods.

Structural econometrics is a field where PhD students, in their 5 year of study, usually produce only one complete study, if that.

Reply on news.ycombinator.com