Jim Simons Trashes Trend Following

at anyrate, if you actually watched the video, he never "Trashed" trend trading, he simply said his math no longer worked on the kind of trend trading he made a lot of money from. And that led him to other applications for the math and new ways of using math for all kinds of problem solving.
 
If price was truly random walk, stock quotes would jump from a dollar, to thousands of dollars to pennies back to millions or billions, in a matter of seconds. The fact that prices behave sequentially and reflect the overall state of the economy is proof enough there's no random walk. As soon as the rw crowd put constraints on their "random" data models (to mimic real price behavior) their data is no longer random. Duh.
 
achilles28 said:
If price was truly random walk, stock quotes would jump from a dollar, to thousands of dollars to pennies back to millions or billions, in a matter of seconds. The fact that prices behave sequentially and reflect the overall state of the economy is proof enough there's no random walk. As soon as the rw crowd put constraints on their "random" data models (to mimic real price behavior) their data is no longer random. Duh.
More...

Prices are not random walk. They are almost a GMB. Take a look at this excellent blog by a young talent.

The only reason you do not get large jumps is because of rules imposed by exchanges. Occasionally you do get them when liquidity dries up, ex. flash crash.

I agree with you over the long term. If you look at monthly data, there is a trend.

Michael Harris has a new blog where he says that trend-followers make forecasts. His blogs are ranked at the top by readers of the Quantocracy website. The blog on trend-following is worth reading. In his new book Michael Harris attacks backtesting, chart analysis, data-mining, trend-following, daytrading. I will start reading it soon when I get a chance.

My conclusion is that Jim Simons is correct, couldn't be otherwise. He made money one way or the other because he understood the game.
 
goodgoing said:
Prices are not random walk. They are almost a GMB. Take a look at this excellent blog by a young talent.

The only reason you do not get large jumps is because of rules imposed by exchanges. Occasionally you do get them when liquidity dries up, ex. flash crash.

I agree with you over the long term. If you look at monthly data, there is a trend.

Michael Harris has a new blog where he says that trend-followers make forecasts. His blogs are ranked at the top by readers of the Quantocracy website. The blog on trend-following is worth reading. In his new book Michael Harris attacks backtesting, chart analysis, data-mining, trend-following, daytrading. I will start reading it soon when I get a chance.

My conclusion is that Jim Simons is correct, couldn't be otherwise. He made money one way or the other because he understood the game.
More...

Good stuff. Sounds like someone smart is reading marketsurfer. fooled by TA-- good book

And thanks for pointing out the quantocracy site. Love the tag line-- anecdotal evidence is not welcomed here--- :)
 
"As far as the laws of mathematics refer to reality, they are not certain, and as
far as they are certain, they do not refer to reality.” Albert Einstein

Apparently Simmons has found something that makes math and reality converge to a profitable degree. Anyone have a clue what that math is (thermodynamics, quantum mechanics, DSP, etc?)
 
You must log in or register to reply here.
Back
Top