Home > The Economics Profession > The limits of statistical inference

The limits of statistical inference

from Lars Syll

causationCausality in social sciences — and economics — can never solely be a question of statistical inference. Causality entails more than predictability, and to really in depth explain social phenomena require theory. Analysis of variation — the foundation of all econometrics — can never in itself reveal how these variations are brought about. First when we are able to tie actions, processes or structures to the statistical relations detected, can we say that we are getting at relevant explanations of causation.

For more on these issues — see the chapter “Capturing causality in economics and the limits of statistical inference” in my On the use and misuse of theories and models in economics.

  1. Jeff Z
    May 7, 2015 at 8:12 pm


    For whatever reason, I happen to be very interested in this sort of thing. It may interest you to know that the American Statistical Association has published a piece by Mark van der Laan, along version of which appears at Stats.org, (http://www.stats.org/super-learning-and-the-revolution-in-knowledge/) and a shorter version of which appears in Amstats, the magazine of the American Statistical Association.

    Two key things stood out for me. One is the need to formulate the problem correctly. The second is the need to think about the experimental design so that the investigator can be reasonably sure that the data that comes from experiment actually addresses the question the team wants to ask. van der Laan in the section “The Art of Statistics” describes the way a lot of medical research is carried out, and it reminded me of the way a lot of research in economics seems to be carried out.

    I would be curious to hear what you think about his proposed remedy discussed at the end of the article.

    Needless to say, it reminded me of a number of your posts here at RWER.

  2. May 8, 2015 at 2:20 pm

    In the future, I am convinced that it will be possible to understand virtually the entire economy in Real-Time. This may sound insane at first glance, but it is becoming increasingly apparent that continuous exponential growth in the use, and development of smarter, and smarter computers makes this highly likely whether we like it, or not. More importantly, it could mean that new money could be created electronically, and could be phased into the Economy, notably for certain vital social, and environmental projects. This would not lead to serious inflation as we have a very clear understanding in Real-Time of the kinds of supply, and demand pressures between companies, and consumers, and also, very powerful super flexible controls that would monitor fluctuations in the Free Market Price…..See my project of Transfinancial Economics for further detail. (http://www.p2pfoundation.net/Transfinancial_Economics )

  1. No trackbacks yet.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s