Fully automated futures trading

Hey GAT,

Is it a good use of time to try to build a more accurate trend filter? Say using machine learning techniques and using more factors/features to redue noise/detection? If not why so?

No, a bad use of time IMHO. I'd be surprised if you could get significantly better out of sample performance from the ragtag bunch of simple trend rules I use myself (I've seen people try, and fail, at this).

You're better off diversifying your system to different instruments and different styles of trading.

GAT
 
Let me add that it looked like I closed out at literally the last hour. I bought back 14.3 this morning, VIX is at 17.3 now. Total saving £30k. Thanks everyone!

Still did -6% today, total drawdown 14.5% from hwm last week.
This was insanely lucky!
 
No, a bad use of time IMHO. I'd be surprised if you could get significantly better out of sample performance from the ragtag bunch of simple trend rules I use myself (I've seen people try, and fail, at this).

You're better off diversifying your system to different instruments and different styles of trading.

GAT

Of the different styles of trading, what would you recommend?
 
No, a bad use of time IMHO. I'd be surprised if you could get significantly better out of sample performance from the ragtag bunch of simple trend rules I use myself (I've seen people try, and fail, at this).

You're better off diversifying your system to different instruments and different styles of trading.

GAT

Hi GAT,

I know you mentioned that trading rules that are too fast can be excluded due to costs.
Let's assume that costs are low or near 0. Is there a timescale at which momentum effect is overwhelmed by noise such that it is useless as a trading rule?
 
Back
Top