Absolute courses - page 50

 
Dr.F.: into a stable probabilistic algorithm

I've been watching the topic from the sidelines, but maybe I've missed the stability, at least on the history.

SZZY: myself started with the same graphical TS, and the charts are similar to yours, though in another way calculated, if you draw delta or % deviation of financial instruments relative to each other it will be similar charts, by the way, the so-called"spread trading" in principle the same TS, except that the horizon of investment and the way of opening simultaneous multi-directional positions is different.

 

I do not expect GBPJPY to rise any further, as is clear from the picture. 50/50, it will go anywhere. But it is possible to sell the dollar against the pound. That is, to buy the pound dollar. Well, it's already hanging there. At +15 pips by the way. Waiting.

 
Doctor, you could monitor the account on .com for example, and it would be less hassle for you, and it would be more convenient and informative for people to monitor the results, and there would be more trust and less talk. Of course, it's up to the owner.
 
alexx_v:
Doctor, you could monitor the account on .com for example, and it would be less hassle for you, and it would be more convenient and informative for people to monitor the results, and there would be more trust and less talk. Of course, it's up to the owner.

I agree with you. My first step is to create the minimum of interest by showing a couple dozen of deals and then I will continue on the real account with the investment password. In the next week I may open only a demo, because there are some things that I have not mentioned in this algorithm and that I would like to calculate before the real one. It is possible to terminate opened trades and then open a demo account. I will do so. And within a week I will calculate something and we will open demonstration trades on a small real account.
 

Good night. If you want, and the algorithm is formalized, I can run the strategy in my program with optimized parameters, which is designed for synthetic trading. Something like that.

For, as has already been said here, testing online is a thankless and unpromising business.

 
Dima.A.:

Good night. If you want, and the algorithm is formalized, I can run the strategy in my program with optimized parameters, which is designed for synthetic trading. Something like that.

For, as has already been said here, testing online is a thankless and unpromising business.


I can easily arrange a run in matcad as well. I don't need anything other than files with quotes history. You are categorically wrong about low information content. It's just more interesting, like this, in forum mode. If we have a 1. physically meaningful hypothesis, 2. mathematical implementation, 3. observable online results confirming the hypothesis for about ten or three transactions, then we don't need anything else. There is no need for history. All the more reason to optimize. There are no parameters in the system that require optimization. Well, that is, they can be invented, such as calculation interval length or "Entry thresholds", but it is all from the evil one. It either works or it does not. If it works, it works even without optimization. If it does not work, do not try to fool yourself.
 

I would still optimise the timing of the sessions and the stoploss/teicrofit parameters. At the very least...

 
alexx_v:
Doctor, you would monitor the bill for example, and you less hassle, and people are more convenient and informative to observe the results, and there will be more confidence, and less talk. It's up to you, of course.

Taks, you should not spoil the show with advice.

Any fool can monitor, but like this, with closed eyes on a minefield, shouting - hey nerdy suckers, look how it should be done!

And here you are with the "monitor".

Don't get in the way. No good. Shame on you.

 
Dima.A.:

I would still optimise the timing of the sessions and the stoploss/teicrofit parameters. At the very least...


TP=SL maybe, session time - no, no, and no. But I tell you, what difference does it make to double the depo per day or have 10% of it? What matters is the STABILITY of this system. If it grows even 2% a day, you are a billionaire, because the step function grows VERY FAST. Stability does NOT depend on optimization, only on efficiency. If you have stability, then the efficiency can be regulated by the lot size or the frequency of transactions. There is no need to bother with it.
 
Mischek2:

Tax, don't spoil the show with your advice.

Any fool can monitor, but this way, with closed eyes on a minefield, shouting - hey nerdy suckers, look how it should be done!

And here you are with the "monitor".

Don't get in the way. No good. Shame on you.


Meanwhile, GBPUSD is evolving as predicted...

Like the others:

you could say ALL the others, for 7 pips excluding the CADJPY spread is nothing, random noise. You could consider that there is zero there.

The Eurodollar is especially sharp. For the elastic stresses tend to relax...

ъ

Anyway, we are waiting for either TP or SL.

Reason: