How to price options and beat the markets with 2nd order partial differential equations.

Bostonerimus1

It takes a lot of maths and computing power to find the inefficiencies of the markets...and an understanding of diffusion and String Theory doesn't hurt either. This is a very interesting little documentary.

https://www.youtube.com/watch?v=A5w-dEgIU1M

LHW99

Haven't watched it, but if they have to go into that level of maths, I get the same feel I get from many statistical equations. Good when first invented, but most of the people trying to use it subsequently won't understand the background requirements and will therefore generate unreliable results.

If a good proportion of the population don't understand percentages, they won't get far with standard differential equations, never mind partial differentials.

Bostonerimus1

LHW99 said:

Haven't watched it, but if they have to go into that level of maths, I get the same feel I get from many statistical equations. Good when first invented, but most of the people trying to use it subsequently won't understand the background requirements and will therefore generate unreliable results.

If a good proportion of the population don't understand percentages, they won't get far with standard differential equations, never mind partial differentials.

The documentary does a good job of explaining the concepts and the viewer does not need any specialized knowledge of maths. The second half of the show concentrates on Jim Simons, an MIT mathematician who started the Medallion hedge fund. The fact that his success hasn't been repeated makes me wonder about confirmation bias and there are some tax issues and questions as to why Medallion performed so much better than Simons' other funds. 

Using maths to drag patterns out of large data sets is obviously topical because of AI, but being a child of the 60's and 70's I immediately think of Asimov's psychohistory in the Foundation Trilogy.