« What happens after the math is done? | Main | Bias, variance, random errors and honest mistakes »


Feed You can follow this conversation by subscribing to the comment feed for this post.

John S.

I get your point, but Satyajit Das seems to be saying exactly that about investment science in this podcast.


Prof. Zvi Bodie of Boston University has been saying for a long time that if you invest in stocks, risk increases with time. See his website (www.zvibodie.com).
Also, read here the section on price volatility: http://www.retailinvestor.org/risk.html#terminal



SB: Sure, not all economists agree that stocks deliver in the long run although the majority probably do. I'm making a guess that Mr. O'Leary is someone who believes in the existence of an equity premium in the long run, and if that's true, then he believes the possibility of predicting something in the long run while not being to predict the same thing in the short term. Thanks for the link.

John: Heard the Das podcast, good stuff. Das is talking about the belief that no trader can have a long-term performance above the market return. Siegel, Malkiel and others are talking about the equity premium, which is the long-term performance of the aggregate stock market above return of some other class of assets (say bonds). I think these two beliefs can co-exist.

Jon Peltier

It's a stretch to say that not believing climate scientists equates to not believing economists. Both topics are politically charged, but they are unrelated.


I think this post deserves a more straightforward rendering since Jon and others are reading it the wrong way.

I'm not making a categorical comparison between economists and climate scientists. The reason given by O'Leary for not believing climate science is very specific: it is the apparent paradox that one can predict the value of something in the long run but not in the short run.

However, this is not only not a paradox but in statistics, it is absolutely natural that one has more confidence in long-term trend predictions than in short-term predictions, and especially so if the data series is subject to large natural variations.

I'm appealing to an analogy in finance: I believe most of us (not all) agree that in the long run, in aggregate and with an average starting time, stocks as an asset class will outperform bonds. However, most of us will also agree that it is almost impossible to predict whether next Tuesday stock prices would be up or down. If you agree with this, then you should believe that it is possible to predict long-term temperatures without being able to predict short-term temperatures.

This belief does not need to have anything to do with whether the long-term temperature predictions are correct or not - in other words, the fact that short-term temperature predictions are wrong does not imply that long-term predictions must be wrong.

And I'm writing about this at length because this belongs to the "unintuitive" parts of statistical thinking.

Verify your Comment

Previewing your Comment

This is only a preview. Your comment has not yet been posted.

Your comment could not be posted. Error type:
Your comment has been posted. Post another comment

The letters and numbers you entered did not match the image. Please try again.

As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.

Having trouble reading this image? View an alternate.


Post a comment

Your Information

(Name is required. Email address will not be displayed with the comment.)


Link to Principal Analytics Prep

See our curriculum, instructors. Apply.
Business analytics and data visualization expert. Author and Speaker. Founder of Principal Analytics Prep, MS Applied Analytics at Columbia. See my full bio.

Next Events

May: 2 New York Marketing Association Big Data Workshop, NYC

May: 5 NYPL Analytics Careers Talk, NYC

May: 8 Data Visualization Seminar, Denver, CO

May: 15 Data Visualization Seminar, Cambridge, MA

May: 17 Data Visualization Seminar, Philadelphia, PA

May: 22 Data Visualization Seminar, San Ramon, CA

Past Events

See here

Future Courses (New York)

Summer: Statistical Reasoning & Numbersense, Principal Analytics Prep (4 weeks)

Summer: Applied Analytics Frameworks & Methods, Columbia (6 weeks)

Junk Charts Blog

Link to junkcharts

Graphics design by Amanda Lee


  • only in Big Data