Search results

This generates out of sample data snooping with high probability of spurious correlation generation. See these interesting articles http://eranraviv.com/sample-data-snooping/ http://www.priceactionlab.com/Blog/2015/03/fooled-by-machine-learning-applied-to-trading-algo-development/ Problem...

This generates out of sample data snooping with high probability of spurious correlation generation. See these interesting articles http://eranraviv.com/sample-data-snooping/ http://www.priceactionlab.com/Blog/2015/03/fooled-by-machine-learning-applied-to-trading-algo-development/ Problem...

Don't work?

Basic is friendlier. I don't think this is the issue. The issue is how much time one can sepnd programming in low level language. If your job is to trade then a higher revel language with low level support is the best. Easylanguage, AFL, etc. If you work for an operation that hires programmers...

There are all sorts of algos that fade indicators. MACD was hit a long ago because it was popular and stochastics too.Now there are not many retail traders and the focus is on things like golden cross and head and shoulders to target longer -term position holders and technical fund trading...

Yes, life must be easy. If I need to do everything in Python the game is lost. A trader must concentrate on trading and not on programming.

Very useful. Thanks!

It's the law. The same way foreign banks are enforced to report US citizen accounts.

I thought that if a system cannot be backtested it is probably not used much. Does anyone know any such system? I can suggest a system based on market context. It is too broad to be able to backtest.

Here is an example of what I'm talking about http://www.towerswatson.com/en-GB/Insights/IC-Types/Survey-Research-Results/2012/10/The-irreversibility-of-time-or-why-you-should-not-listen-to-financial-economists

Things may not be that simple.

I think if you provide personal advice you have to register. If some guy you are educating asks you what to do with his position in AAPL and you say sell, then you have to be registered to do that. There was another thread about this not long ago but I can't find right now.

WFA and forward testing results are conditioned on the data sample(s). Example: you apply WFA to a trend-following system. The first optimization sample involves a nice trend but the walk forward sample is a sideways market. The system then fails in the forward sample. There are problems here...

:Learning curve can be up to 5 years.

Thanks Abattia but I do not think your example related well to trading and to the fact that what we are interested is time-averages, not mean values. This is the reason I am saying that I fail to understand Aronson. I will try to make this clear by noting that what we are interested is the way...

This proprietary trading operation I know is tracking and placing orders for more than 200 patterns in 15-minite data (Price Action Lab generated EL code) with no problems at all for at least two years.

I understand but the concept of infinite universes is also strange or more than that. So to avoid criticism about infinite time he resorts to infinite universes? Sounds peculiar. I think (a) Either he wants to say something we do not understand] (b) or he does not understand what he says...

Why? Are you an examiner of ET members?:)

Stocks have an upward historical bias and cannot be compared to zero-sum forex.Forex traders are speculators, not investors. "All these factors taken as a whole make it almost impossible to fail…" The only certainty in forex is failure. I will remind of this blog article. Maybe longer term...

That statement does not make any sense to me. If you throw a coin infinite times it is like throwing the same coin once in infinite universes. I think Aronson confuses stationary with the frequency definition of probability. It does not make any difference whether you will test a rule in...