Obtaining a stationary BP from a price BP - page 12
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
Here, Sorento, as it seems to me, everything is not so simple and it is not always reduced to hackneyed "profit factor and mathematical expectation" that so simply to say "was not impressed".
You're not impressed with this, Sorento?
not yet.
in 864 days.
And here's a "simple" one for analysis. One year.
And with 149 bucks 8,538.00%. And Sharp is better.
But rusticity and taste are voluntary.
Science fiction is science fiction...
Even if it is "nerdy".
;)
With grasn's help (thanks for that) I've started to develop the following idea.
1. Construct a zigzag. Select the parameters so that the zigzag distribution is as close to normal as possible.
2. We subtract protection from the price, and obtain a series which is close to the stationary one.
3. We predict them for 2 steps - the end of the current wave and the next one. Perhaps we can use a clever regression model, for the time being I limit myself to the ordinary statistics.
4. Predict the residuals (I have not decided on the method yet).
5. Optimize forecast by min. GER between current rest ray + residuals forecast and price.
6. We get the optimization result - a trajectory.
If you are interested, colleagues, join us :-)
In the process...By the way, it's an old idea of mine (or maybe it's not mine alone), but I still have not got around to it. Try to make a "stationary" row (how stationary is a different question) as follows. Each segment has (obviously) a middle point. Transform the spatial WP in x and y in such a way (deform it) that this point becomes "zero" (without breaking the vertices). Don't forget to make the phase "dense everywhere", i.e. form not only the vertices, but all the points between them. You should get a sort of sawtooth signal, actually with a zero average. This signal should be much longer than the original one. If you do it right, you can reconstruct the signal in "normal-temporal" area. And try to predict this signal in head-on, let's say AR, or by a tricky method described earlier :o)
It is difficult to obtain a distribution that is as close to normal as possible.
This is accurate, but obtain a transformation of the original price series that has the following properties
It is quite possible, of course, with some acceptable assumptions. To the question "why it is necessary", the answer is very simple and the only one - it is an opportunity to use the worked out (tested) mate-parameter and no more. My imho.
This is accurate, but obtain a transformation of the original price series that has the following properties
Quite possible, of course with some acceptable assumptions. To the question "why do we need it", the answer is very simple and the only one - it's an opportunity to use the tried-and-true (tried-and-true) math and no more. My imho.
Indeed. Let's transform the BP price series into something like a sine wave, make a minion of money on it and transform it back (well, to maintain the natural equilibrium)... the money, however, will also have to be converted... to zero (as it were).
Just kidding.
Hi Sergey.
It seems to me that one way (perhaps the only way) to deal with non-stationarity generating BP is to use adaptive methods in the TS. To do this, the system must retrain no later than the characteristic time of stationarity (if there is no stationarity at all, then trying to outplay the market is pointless in principle).
The problem with adaptive TS is that they are also retrained according to some algorithm, which is built into them, but it may not coincide with the algorithm of market changes. That is, the algorithm of market changes can coincide with the algorithm of TS retraining at some period of time, but then it can "go away". The market doesn't change according to a given algorithm - that's the problem.....
Once the herd instinct approximation theorem is invented, the problem will be solved.
Once the herd instinct approximation theorem is invented, the problem will be solved.
And it has already been "invented" and you are even using it. That's why all (99.9%) mechanical advisors "drain" the deposit.
Judging by the PAMMs on the internet, this is not the case. There is still a percentage of successful ones. And all of them are practically the same mechanics. This tells us something. In my opinion, it is impossible to use herd instinct and any system separately. That is, in fact, you can't trade using an Expert Advisor only without its analytical expertise. Which, by the way, is given by hours of observation of charts and nothing else.
Indeed. Let's convert the BP price into a sine wave of sorts, make a minion of money and then convert it back to a sine wave (to maintain the balance of nature)... the money, however, will also have to be converted... to zero (as it were).
Just kidding.
Yeah, we've got a lot of jokers here.
Hi Sergey.
Hi Sergey. :о) Good to see you. Where have you been? What interesting things have you studied?
It seems to me that one way (probably the only way) to fight non-stationarity of generating VR, is to use adaptive methods in TS. To do this, the system must retrain no later than the characteristic time of stationarity existence (if there is no stationarity at all, then the attempt to replay the market is meaningless in principle).
Wait, let's take it one step at a time. We have a transformation problem of a series, with the properties given quite concretely:
Suppose you are given such a series and told that it has properties (1), (2), (3). For property (3) you are given a transformation mechanism. What criteria would you use to check them?