New trends in technical analysis. - page 6

New comment
 
zoritch:
I will start with Schrodinger's famous cat.
When they say that there is a cat in the box in the superposition "alive" and "dead", they do not mean that there are two cats at once, one of which is definitely alive and the other definitely dead, and which could interact (for example, interfere with each other to fit into the box). There is only one cat in this quantum state.
In mathematical language, this state can be decomposed into a basis of "states with well-defined vitality" with non-zero projections onto the basis states "alive" and "dead". It is like vectors in space that can point not only strictly along coordinate axes, but also somehow obliquely. The states "alive" and "dead" are orthogonal to each other and do not interact, since - conventionally speaking - they correspond to two mutually exclusive observational possibilities...
Quantum physics as an approach is weak for the market and old as a trend. What did this Schrödinger solve? Nothing much. Opened the box, the cat's dead. Or alive. As luck would have it. As long as you don't measure it (open the box, be an observer), you can't say anything about the cat. And the market is even more complicated. If at first, everything seems to be as in quantum physics, until you open - you do not know if you will have a profit or a moose, but then, after you have opened it, it becomes much more complicated - the cat, i.e. the position is either alive or dead, i.e. sometimes in profit, sometimes in loss. When to close the box? Does quantum physics answer this question? That's right.
 
DC2008:

The market is very volatile, so I recommend choosing an optimisation period of no more than 2 months and no less than 2 weeks. But in principle, you have got the idea right.

If you like, a little advice: quantum frequencies for buying and selling should be calculated separately, because their profitable frequencies do not coincide. I.e. you should get two frequency spectrum graphs.

Good luck.

Thank you very much for the advice.

I have been optimising every week for the last month. I calculate frequencies both for buying and selling (I didn't mention that). In real trading I'm already giving priority to lots on trend trades.

I've tried an approach when reverse trades are opened at losing frequencies. (ie at profitable frequency we trade on our TS, and at unprofitable frequency signals are reversed)

With this approach, it is important to select a strategy that can be correctly filtered. For example in the 1 minute strategy, the filtering is not very good. And on H1 the number of trades is 4 times less. As the result we have only 2 deals per week (instead of 8-10 as usually), at the same time the results of such deals are positive.

 
And all this (in a month on M15) is computed in 7 hours on 4 cores on a not at all weak processor. That's a lot. Have you tried to optimise the calculations programmatically?
 
It's a nice theme. It's fun. Everybody's rocking. About frequencies - they are practically firing away. Try to compare your "frequencies" with opening times of different exchanges, news releases and conference times of all sorts of coddling bankers and you'll get "sines and quanta"))).
 
Mathemat:
And all this (per month on M15) is being calculated for 7 hours on 4 cores of a very weak processor. That's a lot. Have you tried to optimize the calculations programmatically?

Yes, that's how long it takes. The programming code has a resource-intensive loop that runs on every tick. That's why it takes so long.

I tried to optimize the code. 7 hours was the best result.

The optimization is much faster in MT5, but it hangs for some reason.

But then again. I need the results of this optimization only for further trading. The time is not of much importance here. The main thing is that I have enough time to pay for the weekend.

 
FION:
Nice theme. Fun. Everybody's burning. About frequencies - practically firing away. Try to compare your "frequencies" with opening times of different exchanges, news releases and conference times of various bankers and you will get "sines and quanta"))).

I turn to Fion, Vitu and Sanjkku.

Quantum frequencies are almost a new trend in market research. You have not succeeded in your research, probably because everyone puts their own meaning in the concept of quantum frequencies.

With quantum frequencies I have no idea how to build any channels at all. Saneeeek tried it. (maximum one can write an indicator that displays frequencies in the form of bars, similar to Volumes indicator)

Quantum frequencies do not take time into account at all. Frequency at the time of opening of the deal and the result of this deal are taken into account. Find any regularity between the opening of trading sessions, and the frequency of the market is impossible. The news is time-dependent, but the frequency is not.

In the first second of the hour the frequency may be one, and in a minute - another, and in 10 minutes - another. And for example if on a classic TS I need to enter at the opening of a candle, in the case of frequency analysis the deal will have to be opened at 7 minutes of an hour.

For example, for a buy deal to enter a little lower than the opening candle (2), a little more profitable than at the opening of the candle (1). In this case the TP is bigger and the SL is smaller.

 
serler2:

Yes, that's how long it takes. The programming code has a resource-intensive loop that runs on every tick. That's why it takes so long.

I tried to optimize the code. 7 hours was the best result.

The optimization is much faster in MT5, but it hangs for some reason.

But then again. I need the results of this optimization only for further trading. The time is not of much importance here. The main thing is that it must settle at the weekend.

The process can be accelerated by 10 times! Besides, I need to make MT4 optimizer draw the spectrum.

Alexei is right, think about changing the algorithm.

 
serler2:

I turn to Fion, Vitu and Sanjkku.

Quantum frequencies are almost a new trend in market research. You have failed in your research, perhaps because everyone puts their own meaning into the concept of quantum frequencies.

...

Why look for a black cat in a dark room when it's not there at all...
 
DC2008:

....... quantum frequencies for buying and selling must be calculated separately, because their profitable frequencies do not coincide. I.e. there should be two frequency spectrum graphs.

In order to consider anything at the level of quanta, one must have quantum consciousness.

You can find enough literature about it (quantum consciousness) on the internet. It is not that simple.

 
DC2008:

You can speed up the process by a factor of 10! In addition, we need to make the spectrum draw by the MT4 optimiser.

Alexey is right, think about changing the algorithm.

I will try it, but I doubt I will be able to achieve significant time compensation.
12345678910111213...18
New comment