From theory to practice - page 768
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
What if you add a heart rate type filter?
Let's say a tick is a heartbeat, at each tick we need to calculate the heart rate for a fixed time t - usually a minute.
Then we calculate the average heart rate for the sample.
Enter a trade if the current heart rate < average (or maybe the other way around, who knows).
What if you add a heart rate type filter?
Let's say a tick is a heartbeat, at each tick we need to calculate the heart rate for a fixed time t - usually a minute.
Then we calculate the average heart rate for the sample.
Enter a trade if the current heart rate < average (or maybe vice versa, who knows).
I wonder who has a heartbeat like that?)
What if you add a heart rate type filter?
Let's say a tick is a heartbeat, at each tick we need to calculate the heart rate for a fixed time t - usually a minute.
Then count the average heart rate over the sample.
To enter a trade if the current pulse < average (or maybe vice versa, who knows)
Your and A_K2's measurements are bad in that by the time you count anything, it's already over. A_K2 is more of a problem because it counts for 24 hours. Any statistic is a story about the past, but as soon as we enter a trade the story about the future starts). Well, in order to enter a trade it would be nice to estimate at least the present, but not the past. Even the stubborn determinists thought so and started from the current state.
I wonder who has a heartbeat like that?)
something's wrong with the topic...has the flow of ideas dried up and crashed into the grayness of existence?
I need to add fuel to that firebox :-)
If having considered statistics of huge periods we come to conclusion "distribution is almost exactly XXX", and on the other hand it's certain that the flux is not quite random, then it's necessary to catch fluctuations.
That is you can take a tick flow, a sample in 1 time, a sample in 3 ... a sample in N and if some characteristic changes excessively and in the same way we have the honor to say "it can explode". :-) only the samples should be straitened to independent and mutually simple, to keep in mind, that correlations up to 3, maximum up to 5 bars/types can be traced.
We will get a mixture of "Hearst" + A_K , but it is close to the truth - the market behaves unnaturally before the "bang".
there's something wrong with the topic...
we can take a tick stream, a sample in 1 time, a sample in 3... a sample in N and if some characteristic changes excessively and uniformly in them we have the honour to declare that "it's about to bang" :-) only the samples should be straitened to independent and mutually simple, to keep in mind, that correlations up to 3, maximum up to 5 bars/types can be traced.
You get a mix of Hearst + A_K , but it's close to the truth - the market behaves unnaturally before it "bangs".
It has legitimately dried up, for it has stumbled on a philosopher's stone - the process memory parameter and its conditions of use (the applicability table).
Without this healing key everything is nonsense, but with it any strategy will work.
Again, the maximum I've found personally is the excess of the current tick distribution.
I am watching it working together with my TS. I have had +8% in 2 days. Let's see what happens next...
In the old days (hmmm... and it's only been a year - like an eternity...), I would literally break the floor with my feet, dancing the Kamarinsky from such unrestrained profits that I have now on my test demo-signal...
And now - the melancholy and anxious expectation: what will happen tomorrow? And in a month? Ugh...
Well, let's keep watching.
It will be interesting, of course, if we manage (as Dirac discovered the positron at the tip of his pen), using mathematics, to calculate the grail. Well, those who do not like to wade in the mathematical maze, they can take the indicator from MT4 standard set and use it to get a signal to make, for example, such an Expert Advisor:
Test from 22.11.17 to 22.11.18
This, of course, is interesting, if we manage (like Dirac who discovered the positron at the tip of his pen) to calculate the grail using mathematics. Well, those who do not like wandering in the mathematical maze, can take the indicator from MT4 standard set and use it to get a signal to make, for example, such an Expert Advisor:
Test from 22.11.17 to 22.11.18
Actually the same income with the same deposit and spread at current intraday volatility
and when working with a single lot on a trend, can and should be obtained not in one year, but in one or two days,
And not for almost half a thousand trades, but ten times less (not more than 40-45).
Actually, the same income with the same deposit and spread with the current intraday volatility
And when working with a single lot on the trend, it can and should be obtained not in one year, but in one or two days,
And not for almost half a thousand trades, but ten times less (not more than 40-45).
Dreams, dreams, where is your sweetness? You can gain, and you can lose. It's not the potential opportunity that is important, but the fact that the Expert Advisor works with the profit and an acceptable drawdown over a long interval. If you learn how to choose those 1-2 days for such profits and weed out the days when there is a loss, then I'll take my hat off to you).
Dreams, dreams, where is your sweetness? You can gain and you can lose. It is not the potential opportunity that is important, it is the fact that on a long interval the Expert Advisor works with profit. If you learn to choose these 1-2 days for such a profit and eliminate the days when there is a loss, then I take my hat off to you).
Hats off to you for knowing what it takes to do this and being quite capable of doing it.