Jump to content

Mauboussin: Sharpening Your Forecasting Skills


Recommended Posts

About halfway through it, very interesting so far. Not too surprising that better-than-average forecasters share these specific personality traits. Kind of funny the inverse correlation they found between fame and forecasting accuracy--CNBC needs to do a segment on these findings  ;)


Have you read the book they keep referencing, Superforecasting?

Link to comment
Share on other sites

Interesting paper. Some good things to keep in mind when forecasting investing results.


I'll throw out some issues that the paper does not address.


- Paper suggests that short term predictions are easier to make or train to make. Couple questions based on this:

  - is this true in investing?

  - is this true for interesting questions in investing? (Not interesting question might be "Will stock price of A be within 10% of current price after six months?")

  - how does this reconcile with long term views (and long term questions). Clearly a lot of Buffett disciples would rather take a long view. But are long forecasts possible and trainable as short forecasts are? (Clearly if you make 5 year forecasts, you will only know results in 5 years, so your training feedback loop will be very slow).


- Predicting for GJP is quite different from predicting in investing:

  - you can pick your own questions. Which may be positive or negative.

  - you can pick easy questions sometimes (Buffett's one foot hurdle)

  - you may be forced to answer difficult questions if you want to outperform an absolute (10% year) or relative (index) hurdle.


- Measurement and evaluation

  - Measurement is needed for feedback loop and evaluation of your skillz. Is it an issue that you measure yourself (unlike GJP where someone else measured you and classified you as superforecaster, forecaster or a loafer. What is a good measurement for investor forecaster?

  - Assuming you measure, when do you evaluate yourself as superforecaster, forecaster or a loafer. What do you do if your evaluation shows that you're not a superforecaster (yes, training is possible, but some of the qualities mentioned in the paper are nature and not nurture).


Anyway, these are the things that come to my mind now.

I should read the book for some answers perhaps. ;)

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
  • Create New...