Calling All Boglehead Quants-Techies;"Dragon-Kings"

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
User avatar
goodenyou
Posts: 1270
Joined: Sun Jan 31, 2010 11:57 pm
Location: Skating to Where the Puck is Going to Be..or on the golf course

Calling All Boglehead Quants-Techies;"Dragon-Kings"

Post by goodenyou » Sat Jan 27, 2018 3:58 pm

I stumbled upon this by coincidence. A close friend from my past became a very prominent and successful banker/fund manager after he completed MIT in electrical engineering. He has resurfaced with a new company in Switzerland that is all about managing risk. What is interesting is that his partner, French Physicist Didier Sornette, gave a TED Talk about predicting the next financial crisis in 2013 (link below). I found this very interesting. Anyone heard of this? Comments?


https://www.ted.com/talks/didier_sornet ... ial_crisis
"Ignorance more frequently begets confidence than does knowledge" | "The best years you have left are the ones you have right now"

jalbert
Posts: 3461
Joined: Fri Apr 10, 2015 12:29 am

Re: Calling All Boglehead Quants-Techies;"Dragon-Kings"

Post by jalbert » Sat Jan 27, 2018 4:42 pm

The idea that all the information needed to predict the future asset behavior is contained in the past asset behavior ignores the contribution of future unpredictable events that have substantial impact.

The authors do a linear regression of the magnitude and frequency of market turmoil events and talk about outliers. This is just a straw man set up to be knocked down. Whoever said these need to be linearly related?

The authors are correct that extreme market events often exceed what was predicted by institutional risk models. I think this is because the models make liquidity assumptions that causes the models to underestimate risk when the assumptions get violated.

The proposal is a form of technical analysis but wearing non-linear dynamics as a fancy costume.
Index fund investor since 1987.

Post Reply