Neural Networks don't work for trading

[Hugin]

Quote from RedRat:

Thank you Hugin,
I am familiar with GA. But unfortunately my basic NN has 16 input neurons + hidden layer, say 24 neurons, so I suppose it will take enormous time to train with GA . One output.

How many neurons or connections do you have?

I am also trying SVM and RVM but do not have results so far.

Are you familiar with reinforcement learning? Is it better then GA optimization?

RR

Actually the method we use could be classified as reinforcement learning. Most (?) evolutionary algorithms can use the concept of reinforcement learning to explore the state space of the model (if I remember correctly this was included in Holland’s work on GA).

We have looked at SVM but haven’t tried it. The hope is that kernel regression could reduce computation times. I have no experience of RVM.

The total size our network(s) is similar to yours. Our model also includes a number of other parameters not included in the NN.

/Hugin

 

[Hugin]

Quote from TSGannGalt:

Because we're now including GA, in this discussion...

Want to talk about Markov Chains Analysis?

Markov Chains as a model for how GA works or used as a trading model?

/Hugin

 

[Humpy]

Here are some stats from an NN of S&P 500 for last 18 months for comparison

 

[TSGannGalt]

Quote from Hugin:

Markov Chains as a model for how GA works or used as a trading model?

/Hugin

How do you set your Generation and Population size?

 

[Hugin]

Quote from TSGannGalt:

How do you set your Generation and Population size?

Don’t want to turn this thread into a GA specific thread, but here we go. We started with population sizes recommended by Grefenstette (i.e. pretty small ~30) but found that we had to increase it a bit, possibly due to the fact that our problem differs from standard benchmark problems.
We also found papers by Goldberg by that pointed to a population size of ~60. This estimate changes a little depending of what type of model we’re optimizing but a size around 80 seems to be working well.
Regarding the number of generations we set it fairly high (>100) and break a run when the population has not improved over the last 10 generations, as measured by change in average fitness and the diversity of the population. We also have a stop criteria if the elite portion of the population has not changed in the last 10 generations.

/Hugin

 

[TSGannGalt]

Quote from Hugin:

Don’t want to turn this thread into a GA specific thread, but here we go. We started with population sizes recommended by Grefenstette (i.e. pretty small ~30) but found that we had to increase it a bit, possibly due to the fact that our problem differs from standard benchmark problems.
We also found papers by Goldberg by that pointed to a population size of ~60. This estimate changes a little depending of what type of model we’re optimizing but a size around 80 seems to be working well.
Regarding the number of generations we set it fairly high (>100) and break a run when the population has not improved over the last 10 generations, as measured by change in average fitness and the diversity of the population. We also have a stop criteria if the elite portion of the population has not changed in the last 10 generations.

/Hugin

There are better ways to do it.

Just Google "Markov Genetic Algo", you'll find papers regarding them...

 

[Hugin]

Quote from TSGannGalt:

There are better ways to do it.

Just Google "Markov Genetic Algo", you'll find papers regarding them...

Thanks, will do if time permits. The Grefenstette study, even if old, is interesting since they use a meta level GA to optimize GA parameters over a number of benchmark problems. The Goldberg study is based on schema theory.
/Hugin

 

[Hugin]

Quote from TSGannGalt:

There are better ways to do it.

Just Google "Markov Genetic Algo", you'll find papers regarding them...

Did a quick search and found a few articles. Even found a few on island/migration GA. From these it seems that the Markov chain analysis is mostly directed towards population convergence (most stipulate population to be ”large enough” or use the population size as an input to the analysis) and not how to determine a poulation size, but maybe I haven't found the right papers.
Another thing is that we do not use binary string genomes which raises a question. How will this impact the transition matrix in the Markov chain analysis, especially with respect to mutation? Anyway, I found some good pointers on convergence that might become useful.
/Hugin