Fully automated futures trading

nemo4242 said:
About forecast scalars:

I am using artificial random data to estimate the scalars and correlations between rules.
For example, I created 1000 different random walks, each 100000 bars('days') long and applied trading rules to them:

Code: 
                                count  mean  std  min  25%  50%  75%  max
scalar_ewmac256  1000  1.628219  0.070967  1.420118  1.581897  1.626521  1.678534  1.827633
scalar_ewmac128  1000  2.301430  0.071873  2.092129  2.253581  2.300945  2.349642  2.524300
scalar_ewmac64   1000  3.262073  0.073323  3.039892  3.211492  3.261055  3.313205  3.511254
scalar_ewmac32   1000  4.639525  0.075807  4.417496  4.589838  4.636334  4.694308  4.875523

The 50% value is the median along the 1000 tries (data is from the pandas describe() function).
ewmac256 means the EWMA 256/64 rule and so on.
The medians are lower than the values that you give in your book (1.87, 2.65, 3.75, 5.3), probably reflecting the fact that the artificial data has no trends.
But even with 1000 tries of 100000 days, i.e. about 400 years, you can see the error in the values is still significant.
I wonder if it is really possible to get stable values from real data where we only have maybe 40 futures and 50 years of data.
More...

You're right - it isn't possible. Great experiment by the way.

I see a lot of people getting really tied up about the 'correct' value of these things. But it doesn't matter as much as people think. For example if you apply random noise to all your forecasts scalars, such that they're only right within an order of magnitude, what effect would it have on your Sharpe Ratio? Not as much as you might expect.

GAT
 
jj1111 said:
What're you going to spend all that loot on?
More...

Tax take is about a quarter (it's classed as capital gains, which attracts a marginal rate of 28%).

Some goes towards living costs; but I've got enough other investment income to cover most of that.

So most of it is ploughed into buying other investments; firstly through funding ISA allowances (if you're not UK based this is a tax sheltered investment scheme), then buying stuff outside tax shelters. The plan being to have my passive investment income outside of tax shelters increase to the point where it's covering wouldn't matter if my trading account earned zero.

Once I've got that, with a reasonable amount of safety, I might consider paying off my mortgage or buying another place. Only then will I even think about the usual mid life crisis stuff - cars, bikes and boats.

These days are behind me (from 1:30 onwards)



GAT
 
Hi,

Thanks for the all to info you provide on this subject.
Can you point me to sources for the "theory" of correctly optimizing the parameters of strategies ? By "correctly" I mean:
-avoiding over fitting
-correctly using in sample and out of sample data
-resting on solid statistical models (esp. w.r.t. no over fitting).
Thank you.
 
softdown said:
Hi,

Thanks for the all to info you provide on this subject.
Can you point me to sources for the "theory" of correctly optimizing the parameters of strategies ? By "correctly" I mean:
-avoiding over fitting
-correctly using in sample and out of sample data
-resting on solid statistical models (esp. w.r.t. no over fitting).
Thank you.
More...

I'm assuming you've read my book already :-)

If you need more detail then is probably the best book specific to trading.

If you're really serious about it then is the best book in terms of giving you an understanding of distribution of sample statistics.

GAT
 
You must log in or register to reply here.
Back
Top