On the unequal probability of a price move up or down - page 68
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
In principle, the problems with entry accuracy are the same as when trading a single instrument.
It is strange that it is understood by a very limited number of participants.
The only advantage is that the probability of convergence is higher after a large spread.
I wonder how you determined this probability. If it is really higher (which I doubt), then at the time of this large splitting you can open positions in spread with equal virtual SL and TP and with the passage of time on the account will be formed profit. By the way, you can check it in the MT5 tester.
I would call this phenomenon more sluggish spread movement. The plus here is that you can lose less because the spread is more lazy. But the earning opportunities are also much lower in comparison to one pair. It is possible to make a synthetic currency out of all currencies, the exchange rate of which will stand still. But the loss will increase due to swaps.
Thanks, that's interesting! And what are the reasons for the increased likelihood of convergence before non convergence? Does the imbalance between currencies increase? What do you think is driving the reversal ?
The probability of convergence is no higher than the probability of divergence. The probability of convergence only increases when the value of the divergence exceeds the historical average value of the divergence. Also, if the instruments are correlated.
The same thing that reverses the trends of individual currencies. The reasons are many. Fundamental news, crowd psychology and many other things. And yes the balance between currencies is basically always maintained, although black swans do happen.
... Crowd psychology ...
No longer relevant.
I will describe the elements of convergence/divergence in more detail:
1.Looking at XXX/YYYYandXXX/ZZZZ, find the camber
Once we've found it, we look atYYY/YY/ZZZwhere it is located, i.e. we look at the range levels, if it is near to one, we enter the market using this level, it is less likely that the spread will continue. If it continues, then we average.
We look at XXX/YYYandZZZ/PPP, we find the spread - enter on the pair that in our opinion is more inclined to convergence, if the price goes against us - we made a wrong choice, we hedge with the second selected pair and wait 3-5 points and close.
We select pairs that are approximately mirror images, but they do not contain the same currency.If the spread continues, we average the leg that took the lead. 3.
WatchXXX/YYYandZZZ/PPP, wefind the spread - enter both pairs simultaneously and wait for profit.
This method is less preferable, as profit grows more slowly due to the fact that one of the legs gives minus.
In practice I most often use Variant 2. The good thing about it is that you usually make profit much faster than in case of Variant 3
Variant 1 is considered earlier and the screenshots with entry and exit are attached, but hedging is not used there.
I'll continue for system 1 and 2:
There's a chart.
If we enter at 1, we may average so much that the deposit will be insufficient, but if we enter at 2, the reversal is inevitable and we may not even need to average anything.
I am still using this indicator, thanks toVictor Nikolaev.
Forum on trading, automated trading systems and strategy testing
I will write indicator for free
Vitaly Muzichenko, 2018.02.20 21:34
Who is not difficult, please write an indicator:
If the loss from the last opened position exceeds N-points(to output in settings), then the chart background colours are repainted red(to output in settings). If another position is opened, the background will be the default value again until the loss exceeds the N-value in pips. If it is possible, then add a sounder as an alert (enter it into settings).
That's all)
The strange thing is that it is understood by a very limited number of participants.
I wonder how you determined this probability? If it is really higher (which I doubt), then at the time of this large spread you can open positions with equal virtual SL and TP and with the passage of time the account will form a profit. By the way, you can check it in the MT5 tester.
I would call this phenomenon more sluggish spread movement. The plus here is that you can lose less because the spread is more lazy. But the earning opportunities are also much lower in comparison to one pair. It is possible to make a synthetic index out of all currencies, the exchange rate of which will stand still. But the loss will increase due to swaps.
Everything I state is just my opinion. I do not believe I am right about everything I write. I may be wrong or mistaken about some things.
In principle, the problems with entry accuracy are the same as when trading a single instrument.
Grigori.S.B:
The strange thing is that it is understood by a very limited number of participants.
And it is misunderstood. This is clear even from the simplest and most general considerations. Of course, if you want to predict in future a*A + b*B linear combination (where a and b are some constant coefficients and A and B are currency pairs), based on its (linear combination) values in past, then there is no profit, the task is completely equivalent to predicting A or B separately (cases of specific forms of A and B chart are not considered, we mean the typical quotes streams).
But we don't have a*A + b*B as input information (in the past), we have A and B separately. That is MORE information. Do you get it? If there is MORE background information, the prediction can be made FAR more accurate, it's quite obvious. I'm surprised that such elementary physics causes misunderstanding (it is perfectly obvious to sensible people that, for example, from knowledge of A and B, it is easier to predict the difference (A - B), than from knowledge only of A/B (not two relations A/C and B/C, namely only one A/B) to predict the A/B relation).
And if to predict from knowing A and B not some arbitrary "difference (A - B)", but a concrete combination with concrete (convenient) coefficients to A and B, then the prediction accuracy can be even more improve. It's even surprising that such simple things have to be explained. It would seem that many people here have a science or technical education.And misunderstood. This is clear from even the simplest and most general considerations. Of course, if one wants to predict the linear combination of a*A + b*B (where a and b are some constant coefficients and A and B are currency pairs) based on its (linear combination) values in the past, then there is no gain, the task is completely equivalent to predicting A or B separately (cases of specific forms of A and B charts are not considered, we mean typical quote flows).
But our initial information (in the past) is not a*A + b*B, but separately taken A and B. That is MORE information. Do you get it? If the initial information is MORE, the prediction can be made FAR more accurate, it's quite obvious. I'm surprised that such elementary physics causes misunderstanding (it is perfectly obvious to sensible people that, for example, it is easier to predict the difference (A - B) from knowing A and B than to predict the A/B ratio alone (not two A/C and B/C ratios, but just one A/B).
And if to predict from knowing A and B not some arbitrary "difference (A - B)", but a concrete combination with concrete (convenient) coefficients to A and B, then the prediction accuracy can be even more improve. It's even surprising that such simple things have to be explained. It would seem that many people here have a science or technical education.Michael, you are talking nonsense again. You have events or constants appearing as A and B. In the first case, it is closer to Boolean algebra or theorem, in the second - to arithmetic. But financial markets are neither. The price is a very complex function of time and a large number of unknowns. And it is sheer nonsense to operate with A/B/s the way you do. You're not a stupid man and cannot fail to understand this.
It's time to show trick number 7, we will change the pairs from EURUSD and GBPUSD to something else, as requested by the public. For example, USDJPY and EURJPY.
Everything is the same, except that I will draw additional curves in green. As of 19:45 Moscow time on 21.01.2020:
"In one picture it looks like this:
What do we see? That we should buy USDJPY (black curve) and sell EURJPY (red curve), though the divergence is small (33 parrots), so it is very likely that it will increase first (then we will fill it, some would claim it is averaging).
Volume ratio: 2.175.