Statistical modeling as modern astrology

I’m guessing that future generations will look back at our fondness for statistical models in the same way we look back on people who believed in astrology.

It looks like European airports may have been mistakenly shut down because of computer models. The other obvious examples of model failure are the financial crisis and global warming.

The commonality between these events is that they are really complicated. There are way too many variables to control for. The resulting conclusions that come from the models are gibberish as the error bounds are enormous.

Eventually, we may get better at predicting really complex phenomenon, but we’re not good now.

One Response to Statistical modeling as modern astrology

  1. aretae says:

    I’m with you about 90%, and post frequently on the topic, as does the even more experienced Arnold Kling.

    HOWEVER, there’s the problem that there are some systems (Nuclear weapons, Nuclear Power, Airplane wings), where exactly that kind of statistical modeling is used, and you get great results out of it (one of my favorite math professors in school had done his Ph.D. modeling nuclear power reactions). So…you need some sort of refinement in your criticism. Some math modeling works. Just not in complex (many-variable) dynamic (or adaptive) systems like Weather and Macro.

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 )

Google photo

You are commenting using your Google 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 )

Connecting to %s

%d bloggers like this: