Principles of working with an optimiser and basic ways of avoiding fitting in. - page 4
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
Rather the opposite, imho. The stronger the trend, the more inefficient the market, again, imho.
What does stationarity have to do with it though :)
Still, I wonder if the trend is a stationary piece or not?
not all. Stationarity is the self-similarity of the statistical characteristics. I.e. if this trend is broken down into separate chunks, they should have approximately the same statistical characteristics as the entire trend. Of course, the question is about the size of these chunks - they should be statistically significant in size.
I.e. a uniform trend with the variance plotted. The ideal is a straight line at an angle with small fluctuations around it :)
P.S. or a horizontal straight line (flat)
P.S2 it is a question of distribution of price increments on the trend segment, and not of this series
Adaptivity itself and it does not solve the problem of non-stationarity. There are a number of techniques and methods for modelling non-stationarity. As a result, at a minimum, it is possible to reduce the spread of the unsteady residual.
All these methods are shamanism, just like TA without understanding why it should work on a range of prices and not temperatures for example))
not all. Stationarity is a self-similarity of the statistic. I.e. if this trend is broken down into separate chunks, they should have approximately the same statistical characteristics as the entire trend. Of course, it is a question of the size of these chunks - they should be of statistically significant size.
I.e. a uniform trend with a constructed variance. The ideal is a straight line at an angle with slight fluctuations around it :)
P.S. or a horizontal straight line (flat)
But then it still turns out that the non-stationarity kind of decreases during the trend?
But in any case it turns out that non-stationarity seems to decrease during a trend, doesn't it?
trend is a certain MO in price increments (up trend is positive, down trend is negative). But stationarity means keeping the value of this MO over time. I.e. if every hour we steadily increase on average 30 points +-slightly, then the value of the MO varies within small limits, we can say quasi-stationarity. The series +30,+28,+32,+29.... can be considered quasi-stationary after the fact)) But +30,+50,+100,+150... no longer. Because there is no convergence to the mean value as it was in the first case to +30
It has little to do with trading, more to do with analysis of trade results or tests. imha
This is a defeatist position.
Why not assume that unsteady series = sum of several components. And the most interesting one is the deterministic component. If it does not exist or we admit it, then it is a random walk and forecasting is not possible by any means and methods (efficient market theory). If we acknowledge it, then our presence on the market and in this forum is justified.
I agree with you completely and completely disagree (excuse the pun) I will explain: let's take your statement as an axiom and assume that it is a priori (before experience, excuse the free interpretation) - true. In this case, our task is to find a deterministic component and on its basis build a model which will give us mo within the limits of our needs. We find it (extrapolate, use Ito and necessarily Stratonovich - we exactly use it, write a neural network or find regularities expressed as a difference between two averages IMHO much more convenient than Stratonovich and other stochastic dances, the meaning is the same anyway) - who has enough imagination and what is closer to whom. Now we have a function, which, as we have defined, is deterministic (note that we set up an experiment and now assert that it is deterministic a posteriori). Everything in our logic is fine: an a priori model is accepted and on its basis we construct an a posteriori function that extracts the deterministic a posteriori regularities. The only thing is that the patterns we derive are a posteriori. We can never, ever, ever do anything about it. Our task is to find an algorithm that (if we take your assumption as an axiom) will be dynamic-varying with our data in real time(because any deterministic non-stationarity is also non-stationary).
Offtop: about half a year ago I asked on the forum what is the difference between Kiwi (New Zealand) and all other pairs, I managed to obtain some model that gives very good profit (again, a posteriori) on all pairs except Kiwi. No fitting, no Ito-Stratonovich, no self-defeating. Exclusively opening prices, no optimisations. The model is surprisingly simple and straightforward based on the simplest candlestick statistics(which cannot work in the marketin principle - this is what surprised me), moreover random patterns generated also bring profit. But the commodity currencies and the currencies of small economies (those subject to trends) were completely out of the allegedly found pattern. That is the only reason I agree with you - we can admit that there is some determinism, even though it contradicts common sense (excuse the pun), it is this determinism that sometimes prevents ... All this is purely IMHO without claiming to be true.
Rather the opposite, imho. The stronger the trend, the more inefficient the market, again, imho.
Somehow totally support your IMHO.
trend is a certain MO in price increments (up trend is positive, down trend is negative). But stationarity means preserving the value of this MO over time. I.e. if every hour we steadily increase on average 30 points +-slightly, then the value of the MO varies within small limits, we can say quasi-stationarity. The series +30,+28,+32,+29.... can be considered quasi-stationary after the fact)) But +30,+50,+100,+150... no longer. Because there is no convergence to the mean value as it was in the first case to +30
It has little to do with trading, more to do with analysis of trade results or tests. imha
Gee whiz. Have you heard anything about exponential trends?
The smoothing function may have any analytical form. Why should we limit ourselves? And the fact that it is a curve does not change anything, because we have heard of derivatives and know how to handle any smooth curve at any point.
Well, well. Have you heard anything about exponential trends?
The smoothing function can be of any analytical form. Why should we limit ourselves? And the fact that it is a curve does not change anything, because we have heard of derivatives and know how to handle any smooth curve at any point.
I didn't write anything about smoothing functions, trend models, series approximations, etc. It was about stationarity. In particular about the time invariance of the first moment of the distribution (mo). The unit root test you use contains "any kind of smoothing functions", "exponential trends"? :)
Still, I wonder if the trend is a stationary piece or not?