Not the Grail, just a regular one - Bablokos!!! - page 64
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
I saw it, just didn't understand how the synthetic was calculated.
The main question on the synthetics is. Where's the proof that it will come back from the channel boundaries? Maybe it's just luck?
Maybe it will get lucky. Or maybe not! And the proof ....
Is there a need for it? And does it exist?
But if in 75% or more of cases we reach the profit on two positions in total - isn't it (a kind of) statistical proof of such an approach to trading?
Approximately, during one and a half or two years on one forum, I described in advance (!) on a contest account all (well, almost) my entries/exits like that and obtained profit of +5 to +15% of my deposit every month (with a very rare exception of disqualification)! I summed up results every week.
To not confuse all those present - now I'll give you in a private link. All the stats are available there (first months in the first post, all the rest - in the branch).
Maybe you'll get lucky. Or maybe not! And proof ....
Is there a need for it? Does it really exist?
But if in 75% or more of cases we reach the profit on two positions in total - isn't it (a kind of) statistical proof of perspective of such an approach to trading?
Approximately, during one and a half or two years on one forum, I described in advance (!) on a contest account all (well, almost) my entries/exits like that and obtained profit of +5 to +15% of my deposit every month (with a very rare exception of disqualification)! I summed up results every week.
To not confuse all those present - now I'll give you in a private link. All the stats are available there (the first months in the first post, all the rest - in the main thread).
The proof is there.
We can also take the balance as proof, but it is faith and if we believe, we trade - all TA is built on faith.
It's all clear to me here.
Here is a picture
At point "A" there will be the largest divergence and we will enter into a byu-sell position. But we see that both instruments are in a rising trend, just the higher one overtakes the lower one. We will make a profit if the loss on one instrument outweighs the profit on the other.
What confuses me in this whole idea is the fact that we are entering against the trend in one of the instruments. We are not taking into account the overall trend of the pairs going together, in your terminology, correlated.
The proof is there.
We can also take the balance as proof, but it is faith and if we believe, we trade - all TA is built on faith.
It is clear to me here.
Here is a picture
At point "A" there will be the largest divergence and we will enter into a byu-sell position. But we see that both instruments are in a rising trend, just the higher one overtakes the lower one. We will make a profit if the loss on one instrument outweighs the profit on the other.
What confuses me in this whole idea is the fact that we are entering against the trend in one of the instruments. We are not taking into account the overall trend of the pairs going together, in your terminology, correlated.
And don't be embarrassed, be brave - if you choose the right shares of instruments in the portfolio, the overall total will be profitable.
=============
It is also possible to "dabble" in statistical arbitrage. The graph below is an example of a synthetic all-aluminium vs. all-aluminium tool.
I.e. NA -(ZN+HG+NI), alluminium - (zinc+copper+nickel)
The channel is a little slanted, but quite comfortable for short-term (and even automatic) work:
Well here - "variations are possible"...
We take at a certain moment of time the symbol whose price line (the upper indicator) has deviated from all others! We set the lower indicator to draw such a synthetic. And on the reversal - work with this tool - "against all" others!
For example, yesterday evening on m15 it was possible to find an entrance "copper vs. all others", i.e.
BUY 3*0.28*HG - (SELL 1*AL + SELL 0.97*ZS + SELL 0.28*NI)
=============
It is also possible to "dabble" in statistical arbitrage. The graph below is an example of a synthetic all-aluminium vs. all-aluminium tool.
I.e. NA -(ZN+HG+NI), alluminium - (zinc+copper+nickel)
The channel is a little slanted, but quite comfortable for short-term (and even automatic) work:
Well here - "variations are possible"... We take at a certain moment of time the symbol whose price line (the upper indicator) has deviated from all others! We set the lower indicator to draw such a synthetic. And on the reversal - work with this tool - "against all" others!
For example, yesterday evening on m15 it was possible to find an entrance "copper vs. all others", i.e.
BUY 3*0.28*HG - (SELL 1*AL + SELL 0.97*ZS + SELL 0.28*NI)
I don't understand anything in your thread or here.
Above you wrote that you took the difference of two instruments. And here?
Where do the values of the ratios come from?
With what error are these values calculated?
How long do they live? Maybe on the next bar, when you are in the pose, these coefficients will have a different value?
Above, you wrote that you took the difference of the two instruments. And here?
Where do the coefficient values come from?
With what error are these values calculated?
How long do they live? Maybe on the next bar, when you are in the pose, these coefficients will have a different value?
faa1947, if you have understood the 2 tools, then I think you can do the same here.
Here I took not two but four tools. In other words, we enter/exit the market using four positions simultaneously. Two in buy + two in sell. Or one in buy + three in sell. Or one in the sell + three in the buy.
Coefficients (i.e. sizes of positions, or rather their ratio) are calculated by the upper indicator taking into account instrument specifications. And to the right of the name - these ratios are also calculated taking into account the volatility of each instrument (1-0.97-0.28-0.1).
On the next bar (and on another timeframe as well) they may change. But only slightly - at least I usually round these figures to a comfortable trading value.
===============
We can see that the copper price and prices of other instruments have diverged and are only just beginning to converge. At this point (vertical yellow line) we make an entry:
BUY 3*0.28 *HG - (SELL 1*AL + SELL 0.97*ZS + SELL 0.28*NI)
assuming copper rises faster or falls slower than the other three metals.
faa1947, if you have figured out the 2 instruments, I think you will figure it out here too
Here I have taken not two, but four instruments. In other words, here we enter/exit the market with four positions simultaneously. Two in buy + two in sell. Or one in buy + three in sell. Or one in the sell + three in the buy.
Coefficients (i.e. sizes of positions, or rather their ratio) are calculated by the upper indicator, taking into account instrument specifications. And to the right of the name - these ratios are also calculated taking into account the volatility of each instrument (1-0.97-0.28-0.1).
On the next bar (and on another timeframe as well) they may change. But only slightly - at least I usually round these figures to a comfortable trading value.
===============
We can see that the copper price and prices of other instruments have diverged and are only just beginning to converge. At this point (vertical yellow line) we make an entry:
BUY 3*0.28*HG - (SELL 1*AL + SELL 0.97*ZS + SELL 0.28*NI)
assuming that copper will rise faster or fall slower than the other three metals.
Thanks, I think I've got it figured out.
Especially the idea of trading one instrument against several others is interesting. I was having trouble figuring out how to trade a pair if a non-tradable synthetic was against the instrument being traded. Now I understand how to do it.
Thanks again.
faa1947, if you figured out the 2 tools, then I think you will figure it out here as well
Here I have taken not two, but four instruments. In other words, here we enter/exit the market with four positions simultaneously. Two in buy + two in sell. Or one in buy + three in sell. Or one in the sell + three in the buy.
Coefficients (i.e. sizes of positions, or rather their ratio) are calculated by the upper indicator, taking into account the instruments' specifications. And to the right of the name - these ratios are also calculated taking into account the volatility of each instrument (1-0.97-0.28-0.1).
On the next bar (and on another timeframe as well) they may change. But only slightly - at least I usually round these figures to a comfortable trading value.
===============
We can see that the copper price and prices of other instruments have diverged and are only just beginning to converge. At this point (vertical yellow line) we make an entry:
BUY 3*0.28*HG - (SELL 1*AL + SELL 0.97*ZS + SELL 0.28*NI)
assuming that copper will rise faster or fall slower than the other three metals.
By the way your profit results, which were on the link, are these with or without MM?
As far as I understand, it is essentially something like cluster indicators. It does not analyse the cointegration of the price series themselves, but is based on price returning to the moving average. I.e. we take deviations from МА which are known to be a stationary process. That is why in principle we may take any coefficients, for any linear combination of stationary processes is a stationary process. Naturally, the profit is not guaranteed here because either the mountain goes to Mohammed, or Mohammed goes to the mountain :)
In general, the question about the profitability of this system is ambiguous. Watching the work of Leonid, I often noticed that he often closed positions without waiting for a signal (conjunction of lines), based on instinct :). Therefore, it is not certain that automated trading will prove to be profitable :)
I.e. it takes deviations from the MA, which is known to be a stationary process.
I don't know from where, could you please indicate the source of such information?