Not financial advice. For testing purposes only. Do not use on live account
Predictive Analysis & Forecast 118 replies
Spud's Predictive MTF Stochastics 1,695 replies
Non-Predictive Trading 53 replies
DislikedOkay, assume we have a very fast neural network that can analyze x (where x can be large) amounts of data very fast, ready to trade the coming set of bars (only thinking 1-3 bars, not more). What would be the best predictive algo to feed such neural network with? "Perceptron" (various implementations online) is a reasonably good one, but there is likely better. Which one do you think is best? Why?Ignored
Disliked{quote} Very nice! Reading up on Wavelets. If you have any work done on this already, shoot me a PM and we can join forces.Ignored
QuoteDislikedIt turns out that there are many ways to do so, but no best one.
QuoteDislikedThis lack of universally best choice can actually be formalized in what is
called the No Free Lunch theorem, which in essence says that, if there is no
assumption on how the past (i.e. training data) is related to the future (i.e. test
data), prediction is impossible. Even more, if there is no a priori restriction on
the possible phenomena that are expected, it is impossible to generalize and
there is thus no better algorithm (any algorithm would be beaten by another
one on some phenomenon).
QuoteDislikedB No Free Lunch
We can now give a formal definition of consistency and state the core results
about the impossibility of universally good algorithms.