Neural Networks

[Indrionas]

Quote from PredictorX:

None of this is relevant. For a claim that "neural networks can never be used productively for trading" to be proven beyond doubt would require an impossibly exhaustive experimentation. The most one could seriously claim is something like "[assertion]: no examples of neural networks aiding trading exist and [opinion:] none is likely to surface". Whether that assertion is true or not, I cannot say, but superlative claims that neural networks can "never" work are silly.

Neural networks can help in model development, and it has been done. So "never" is false.

I guess those who say NNs don't work either don't know how to use them, or tried to use them with wrong data and/or incorrectly. NN is just a tool and before you can use it you have to understand the math and statistics behind them, at least in abstract fashion.

 

[PredictorX]

Quote from SnowMafia:

Aren't genetic algorithms in many instances used in conjunction with neural nets. I believe genetic algorithms are used to train neural nets or to evolve the best neural net available.

While neural networks can be trained by genetic algorithms, they are much more often trained by some other, more direct procedure (gradient descent, for instance). The important thing is to understand that genetic algorithms, in themselves, do not "learn" anything- they are optimizers of other systems. To say that genetic algorithms are "better" some how than neural networks is like saying that saws are "better" than hammers!

 

[Pathfinder]

Genetic algorithms can also simply build neural networks. In that case neural network structure does not depend on generic algorithm but on sequence of examples that were used by the generic algorithm during training.