Machine learning in trading: theory, models, practice and algo-trading - page 3169
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
This is the kind of thing that happens. On the left OOS passes, on the right - not. And on the right side, it literally "dives" immediately.
It happens most of the time.
I.e. literally immediately significant dive. The nature of the dive is not clear. I think it should be something close to SB, but I see such a picture too often.
It feels like if after optimisation you run an inverted TS, you may not even drain.
The tester is worthless in general, even Renat admitted that there will be a new one, the question is when?
A couple of days ago I was looking for my topic there, and came across 2019, someone wrote there that the tester works only on standard indicators.
it's really true, they have added so much functionality that the tester can't cope with the enthusiasm of the users.the tester is worthless
It is solved by replacing the gasket like in a car.
If like the resistance is hotter because the current is high. The new factors of the price model are not taken into account, the model has changed critically.))))))
I have come across situations where such a "worthless" set showed great situations after a few months.
As if there is a switch on the market to switch on/off long term stable market patterns.
It is solved by replacing the gasket as in a car.
teach me how to use it first.
Sub, I am personally sure that 99% here will not be able to test correctly, the result of the tester will be different, I hope you saw my thread recently.
teach me how to use it first.
Sub, I am personally sure that 99% here will not be able to test correctly, the result of the tester will be different, I hope you saw my thread recently.
This thread is not about the tester.
This thread isn't about the tester.
When it's a serious work issue, you can get in,
stop being tetanus.
Faced situations where such a "worthless" set showed great situations a few months later.
It is as if the market has a switch to switch on/off long-term stable market patterns.
Well, fixing the change of pattern at the moment of this change is not a solvable task yet, only by fact. In your case, one of the signs of pattern change is this chart. Perhaps it is possible to find a model for plum plots, or somehow mark them and switch them off from trading, but the beginning of the plot that does not correspond to the profitable TS model has no solution yet.
If suddenly there is some other sign, except for the chart, of model conformity, it is already a good help)))))
This is the picture seen by almost every user of the tester. I am interested in the explanation.
This is the picture seen by almost every user of the tester. I am interested in the explanation.
In this picture, the statistical significance is quite high: more than 3000 non-overlapping positions.
I assume that this is the effect of market changes within Sample itself. For example, Sample had a real pattern in the beginning and then nothing. But the fitting happened for the whole Sample.
We should somehow avoid such breakdowns within Sample.
The opposite effect can also happen: on the left OOS - down, on the right - up. I.e. no pattern was found in the initial piece of Sample, but only fitting.
Overtraining or non-stationarity, I guess. Unless, of course, there are problems with the raw data or the algorithm itself.
I usually try to "move" the problem a bit - slightly change all possible parameters (and available metaparameters) and see how the result changes. Sometimes it becomes a bit clearer.