3 ms·

From the original post: "Much research in economics and finance uses secret data. The journals publish results and conclusions, but the data and sometimes even

by cs702 11y ago

From the original post: "Much research in economics and finance uses secret data. The journals publish results and conclusions, but the data and sometimes even the programs are not available for review or inspection. Replication, even just checking what the author(s) did given their data, is getting harder." If the data and programs remain secret, it's not really science; it's just another example of Richard Feynman's cargo cult science: http://calteches.library.caltech.edu/51/2/CargoCult.htm http://calteches.library.caltech.edu/51/2/CargoCult.htm Sadly, much if not most "economics" research is cargo cult science... often motivated by political, religious, and moralistic ideologies. -- EDIT: In response to davidw's comment below, here are a few other economists weighting on the topic: * http://paulromer.net/wp-content/uploads/2015/05/Mathiness.pdf http://paulromer.net/wp-content/uploads/2015/05/Mathiness.pd... * http://paulromer.net/mathiness/ http://paulromer.net/mathiness/ (shorter version) * http://noahpinionblog.blogspot.com/2015/09/theory-vs-data-in-economics.html http://noahpinionblog.blogspot.com/2015/09/theory-vs-data-in... * http://noahpinionblog.blogspot.com/2013/12/freshwater-vs-saltwater-divides-macro.html http://noahpinionblog.blogspot.com/2013/12/freshwater-vs-sal... * http://news.mit.edu/2010/krugman-event http://news.mit.edu/2010/krugman-event

davidw 11y ago
> Sadly, much if not most "economics" research is cargo cult science... often motivated by political, religious, and moralistic ideologies. Got any data for that, or do you just happen to dislike some of the conclusions? I agree entirely about science without data not being much good, though.
- cs702 11y ago
  The author of the original post, John Cochrane, is a leading economist[1], and he is complaining that "much research in economics and finance uses secret data." How's that for evidence? Did you read his post? In case it wasn't obvious, I agree with him! -- [1] http://faculty.chicagobooth.edu/john.cochrane/ http://faculty.chicagobooth.edu/john.cochrane/
  - davidw 11y ago
    You jumped to conclusions from that, though.
    - fhdjfjcj 11y ago
      The "comments" section exists for the purpose of making "commentary."
    - cs702 11y ago
      No, I din't! I come across this issue repeatedly in the economics literature, but didn't have the inclination to google up more evidence when writing the above comment, as I assumed "everyone knows this." Evidently I was wrong about that. Here are a few other economists (other than Cochrane) weighting on the topic: * http://paulromer.net/wp-content/uploads/2015/05/Mathiness.pdf http://paulromer.net/wp-content/uploads/2015/05/Mathiness.pd... * http://paulromer.net/mathiness/ http://paulromer.net/mathiness/ (shorter version) * http://noahpinionblog.blogspot.com/2015/09/theory-vs-data-in-economics.html http://noahpinionblog.blogspot.com/2015/09/theory-vs-data-in... * http://noahpinionblog.blogspot.com/2013/12/freshwater-vs-saltwater-divides-macro.html http://noahpinionblog.blogspot.com/2013/12/freshwater-vs-sal... * http://news.mit.edu/2010/krugman-event http://news.mit.edu/2010/krugman-event Finally, if this is of interest, I highly recommend you read this opinion piece by Richard Posner, in which he concluded that the dominant conception of economics until the global financial crisis was the abstract-theoretical study of "rational choice" as opposed to the study of how economies actually work: https://newrepublic.com/article/69601/how-i-became-keynesian https://newrepublic.com/article/69601/how-i-became-keynesian If I get the time, I'll google up more blog posts and supporting evidence.
      - davidw 11y ago
        Some of what you point out is that macroeconomics is difficult. Even micro is difficult, because you can't just go running all the experiments on people that you'd like to, but running experiments with entire countries is "kind of" difficult. That doesn't mean it's worthless or mostly partisan or 'religious' (that seems particularly nonsensical). It's fair to point out that economics - like most human fields of study and endeavor - has its problems, but I do object to the broad-brush one-liner style of discussion.
- gone35 11y ago
  Got any data for that, or do you just happen to dislike some of the conclusions? Soo much evidence, you clearly have not been paying attention. See e.g.: [1] http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1683617 http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1683617 [2] https://www.gsb.stanford.edu/sites/default/files/research/documents/Chameleons%20-The%20Misuse%20of%20Theoretical%20Models%20032614.pdf https://www.gsb.stanford.edu/sites/default/files/research/do...
tvanantwerp 11y ago
I work at a think tank full of economists. Most of the data they use is publicly available government data. And I've been working with them to get more and more code into public GitHub repos. Maybe it's different in academia, but we try to be fairly transparent.
- bjelkeman-again 11y ago
  Where is this?

by cs702 11y ago

davidw 11y ago

> Sadly, much if not most "economics" research is cargo cult science... often motivated by political, religious, and moralistic ideologies. Got any data for that, or do you just happen to dislike some of the conclusions? I agree entirely about science without data not being much good, though.

cs702 11y ago
The author of the original post, John Cochrane, is a leading economist[1], and he is complaining that "much research in economics and finance uses secret data." How's that for evidence? Did you read his post? In case it wasn't obvious, I agree with him! -- [1] http://faculty.chicagobooth.edu/john.cochrane/ http://faculty.chicagobooth.edu/john.cochrane/
- davidw 11y ago
  You jumped to conclusions from that, though.
  - fhdjfjcj 11y ago
    The "comments" section exists for the purpose of making "commentary."
  - cs702 11y ago
    No, I din't! I come across this issue repeatedly in the economics literature, but didn't have the inclination to google up more evidence when writing the above comment, as I assumed "everyone knows this." Evidently I was wrong about that. Here are a few other economists (other than Cochrane) weighting on the topic: * http://paulromer.net/wp-content/uploads/2015/05/Mathiness.pdf http://paulromer.net/wp-content/uploads/2015/05/Mathiness.pd... * http://paulromer.net/mathiness/ http://paulromer.net/mathiness/ (shorter version) * http://noahpinionblog.blogspot.com/2015/09/theory-vs-data-in-economics.html http://noahpinionblog.blogspot.com/2015/09/theory-vs-data-in... * http://noahpinionblog.blogspot.com/2013/12/freshwater-vs-saltwater-divides-macro.html http://noahpinionblog.blogspot.com/2013/12/freshwater-vs-sal... * http://news.mit.edu/2010/krugman-event http://news.mit.edu/2010/krugman-event Finally, if this is of interest, I highly recommend you read this opinion piece by Richard Posner, in which he concluded that the dominant conception of economics until the global financial crisis was the abstract-theoretical study of "rational choice" as opposed to the study of how economies actually work: https://newrepublic.com/article/69601/how-i-became-keynesian https://newrepublic.com/article/69601/how-i-became-keynesian If I get the time, I'll google up more blog posts and supporting evidence.
    - davidw 11y ago
      Some of what you point out is that macroeconomics is difficult. Even micro is difficult, because you can't just go running all the experiments on people that you'd like to, but running experiments with entire countries is "kind of" difficult. That doesn't mean it's worthless or mostly partisan or 'religious' (that seems particularly nonsensical). It's fair to point out that economics - like most human fields of study and endeavor - has its problems, but I do object to the broad-brush one-liner style of discussion.
gone35 11y ago
Got any data for that, or do you just happen to dislike some of the conclusions? Soo much evidence, you clearly have not been paying attention. See e.g.: [1] http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1683617 http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1683617 [2] https://www.gsb.stanford.edu/sites/default/files/research/documents/Chameleons%20-The%20Misuse%20of%20Theoretical%20Models%20032614.pdf https://www.gsb.stanford.edu/sites/default/files/research/do...

tvanantwerp 11y ago

I work at a think tank full of economists. Most of the data they use is publicly available government data. And I've been working with them to get more and more code into public GitHub repos. Maybe it's different in academia, but we try to be fairly transparent.

bjelkeman-again 11y ago
Where is this?