Hacker News new | past | comments | ask | show | jobs | submit login

Does anyone else find it strange that there is no real-world data in these notebooks? It's all simulations or abstract problems.

This gives me the sense, personally, that economists aren't interested in making accurate predictions about the world. Other fields would, I think, test their theories against observations.




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


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


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.


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.


There are serious critiques of economic theory out there, which tend to say that kind of thing.

But if you compared these notes to the notes for a college level physics course, you would find a similar level of abstraction, idealized models, and absence of real world data. Those things are not in themselves indicators that physicists (or economists) don't care about the real world. In any mature field, there is a body of knowledge and techniques to be learnt. There's a certain formalism to be picked up, rather than just staring at data.

There might be legitimate reasons for dismissing the general approach taken by mainstream economic theory, but what you seem to be saying ("hmmm, my intuition is that this stuff doesn't focus enough on accurately predicting the real world") is not a reasoned critique.


pandas-datareader can pull data from e.g. FRED, Eurostat, Quandl, World Bank: https://pandas-datareader.readthedocs.io/en/latest/remote_da...

pandaSDMX can pull SDMX data from e.g. ECB, Eurostat, ILO, IMF, OECD, UNSD, UNESCO, World Bank; with requests-cache for caching data requests: https://pandasdmx.readthedocs.io/en/latest/#supported-data-p...

The scikit-learn estimator interface includes a .score() method. "3.3. Model evaluation: quantifying the quality of predictions" https://scikit-learn.org/stable/modules/model_evaluation.htm...

statsmodels also has various functions for statistically testing models: https://www.statsmodels.org/stable/

"latex2sympy parses LaTeX math expressions and converts it into the equivalent SymPy form" and is now merged into SymPy master and callable with sympy.parsing.latex.parse_latex(). It requires antlr-python-runtime to be installed. https://github.com/augustt198/latex2sympy https://github.com/sympy/sympy/pull/13706

IDK what Julia has for economic data retrieval and model scoring / cost functions?


> This gives me the sense, personally, that economists aren't interested in making accurate predictions about the world. Other fields would, I think, test their theories against observations.

You say this as though using mock-up data to teach techniques isn't a universal practice in literally every other discipline.


>Does anyone else find it strange that there is no real-world data in these notebooks? It's all simulations or abstract problems.

Pretty much every course I took in undergrad physics had no real world data. The intro level courses were especially fun, when we'd go into the lab and get such horrible data that we'd never conclude what they're teaching in the theory classes. We wondered what the point of the lab even was.

The biggest offender is the friction model. Heck no - it's not proportional to the normal force. No one could successfully show that in the lab. And a quick Google search shows you a trivial experiment where just changing the orientation and keeping the normal force the same leads to wildly different frictions.


It's an academic course. You learn using models and basic concepts, then eventually apply it to real data.

Ever taken statistics courses? You're not doing multiple regression analysis on real world data on day 1. On day 1 you're learning odds using playing cards and coin flips.


>Ever taken statistics courses? You're not doing multiple regression analysis on real world data on day 1. On day 1 you're learning odds using playing cards and coin flips.

Curiously enough, my undergrad statistics textbook was loaded with problems where the data was taken straight from a journal paper. The book has poor reviews on Amazon, but I think it's the best I've seen.

https://www.amazon.com/Probability-Statistics-Engineering-Sc...


Yeah, higher levels stats had plenty of real world examples. I did real world examples in econ as well. But you build to both.


This wasn't a higher level stats course. It was the first and only course I took (introductory).


That's an ironic conclusion to draw.

You could test your own theory against observations that calculations with real world data are very much a part of economics, but are just not part of this particular course.


And wait until they fit 12 parameters on 20 observations where each of the observations has as well a massive measurement error.


You’re pretty much right on the nose.

Of course, it depends on who they work for. Effectively, the American field of economics is an exercise in decoupling private reality from public theory.




Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact

Search: