Discussing the article: "Fast trading strategy tester in Python using Numba" - page 5
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
Imho, it would be worth to reformulate this opposition in terms of MOE. There are two models far apart on the bias-dispersion trade-off curve. Due to a small fixed number of parameters, the TC has a shift towards increasing bias (linear regression is a common example for MO), while the complex model, on the contrary, has a shift towards increasing variance.
Obviously, if the simpler model captures the actual pattern, it is better. If neither model captures it, then again, the simpler one is better - it is harder to see its fallacy in the complex one because of its better adaptability to noise) Not surprisingly, there is a point to complication only if it is beneficial. This is the obvious theoretical answer.
If a bit more practically, then essentially the second point means stacking models (at least two) - one model breaks down (looking for discrepancies) and the other makes trading decisions. There can also be a third model that switches the trading model on/off, etc. Stacking, as it is known, has a reputation of "black magic" in MO) As a rule, it is used by winners of all kinds of competitions, but there is no theory or recipes for it. If you're lucky enough to find a working stacking, good for you). Imho, stacking simpler models generally makes more sense than trying to cram everything into one more complex one.
Yes, the decomposition problem needs to be solved, since our series are non-stationary. But I wouldn't emphasise it, because it will be solved anyway - either explicitly or implicitly)
So for now we'll call it still magic, since we can't find a more meaningful definition :)
I've come across that staking is a rather tricky construction of several models, not at all similar to what is usually called staking. I don't know how widely accepted that is.
For some reason I'm more comfortable with "stacking", probably because it's a "stack". But let it be "stacking" for the sake of uniformity.
I have encountered that staking is a rather tricky construction of several models, not at all similar to what is usually called staking. I don't know how common it is.
For some reason I'm more comfortable with "stacking", probably because it's a "stack". But let it be "stacking" for the sake of consistency.
Std in a sliding window with different periods, default period is 20. From my phone, I may not see something, I apologise.
Wouldn't it be more correct to read z-score?
Wouldn't it be more correct to consider z-score?
I don't know, what's better?
I don't know, what's better?
It's a normalised value
is the normalised value
Keep going until you get the idea.
Well, the standard deviation in a fixed value sliding window will have a non-normalised range of variation depending on volatility. As far as I know, usually z-score is used for this since it is a normalised value. That's the end of the thought )
Well, the standard deviation in a fixed value sliding window will have a non-normalised range of variation depending on volatility. As far as I know, usually z-score is used for this purpose as it is a normalised value. That's the end of the thought )
Got it, I take min/max over all available history and set as bounds, then split into random ranges at each iteration of the optimiser. You can also do zscore. I thought such normalisation might be better for the optimizer (getting rid of small values with a large number of zeros after the decimal point), but I don't think it should be.