An effective trading strategy based on multi-currency analysis of multiple DCs - page 9

New comment
 
timbo:
Piligrimm:
I hope that the number of people sceptical about multicurrency analysis has decreased, and that the results will help someone find an effective trading strategy.
I think the scepticism was towards multicurrency analysis. It was not intended to criticize the multicurrency analysis, on the contrary, it is being discussed quite effectively in the next thread. And I myself have sinned drawing different charts and calculating correlations of different currencies - I've got something so interesting, that I haven't understood it yet ;-)

I absolutely agree. Multi-learning analysis is a pointless exercise. And there is no normal brokerage company (and probably just a simple "kitchen"), whose quotes, when compared to any other brokerage company, may differ by more than 2 spreads, long enough to open a position. And even if the difference is 3 spreads, using this drawback for stable earnings does not seem possible for certain reasons. And this could be the easiest, most reliable and risk-free way to trade.
As for the multi-currency one...
There are a lot of opinions, but I will only say that I know one way, which allows to make a stable profit in Forex. This variant is not realizable within the framework of one pair, but it is realizable within one brokerage company.
 

I'm now tracking the effects and influences of time intervals between ticks, thinking how best to apply this in multicurrency, at least thanks to these intervals, which I now observe, made some error-free bets, also, I have inserted in the program setting different levels of visualization http://community.xnsnet.ru/photos/mtterm/default.aspx (look in the original screenshot size), based on which I check the efekiness of this or that method in the output of ticks. Now I don't have to keep a close eye on ticks, that's the main thing to understand what and where is flowing:) For example you can watch the time difference with different accuracy, handy for determining flow rate, frost, etc. I think maybe we should put in an average speedometer, just do not know yet how to implement it so that it is really useful, probably the same as with the tick charts, it is necessary to stick the maximum number of settings:) I am waiting at least one full funnel:)

 
xnsnet:

I'm now tracking the effects and influences of time intervals between ticks, thinking how best to apply this in multicurrency, at least thanks to these intervals, which I now observe, made some error-free bets, also, I have inserted in the program setting different visualization levels http://community.xnsnet.ru/photos/mtterm/default.aspx (look in the original screenshot size), based on which I check the efekiness of this or that method in the output of ticks. Now I don't have to keep a close eye on ticks, that's the main thing to understand what and where is flowing:) For example you can watch the time difference with different accuracy, handy for determining flow rate, frost, etc. I think maybe we should put in an average speedometer, just do not know yet how to implement it so that it is really useful, probably the same as with the tick charts, it is necessary to stick the maximum number of settings:) I am waiting at least one full funnel:)


I think the analysis of ticks, more precisely their quality (delta) and frequency, makes sense, although in my case - the higher the timeframe, the more reliable the signal. But one should hardly analyze DC methods (frosts, filtration etc.), because one cannot build a robust TS on it. And if we manage to do it, it will not necessarily mean that brokerage companies will continue filtering and freezing quotes the same way they used to. Therefore, when analyzing ticks patterns formed by movements exceeding spread or more may be useful, if they (patterns) exist, of course, which I have not yet managed to doubt.
 

Boredom in the market now, nothing happens, but in this boredom something happens sooner or later:) Gold, for example, flows back and forth, so it's the little things in life:) Funnels are unfortunately a very rare phenomenon...

In general I don't even think about Expert Advisors, but when I do I will probably not get signals of every phantom, which creates the illusion of movement. I want to make a programmer who will not know what tics are, but will know about signals to actions and maybe about level of errors. I'm more than sure it is possible to write such a machine that will surely make pure profit unhurriedly, without analyzing history in the tester at least, perhaps using both manual and automatic means. Anyway, this is how I see it, for some reason I think it is impossible to write a strong thing based on the same Expert Advisors that will not slow down so much, recalculating at every tick all that I do in one pass. On a multi-currency it would be very noticeable.

 

For example, right now I have detected symptoms of a funnel formation on EURUSD thinking that it may jump in favour of the Euro. Again, it is hard for me to characterize how I did it and how real it is, such symptoms are repeated several times during the day, but it is hard to say for sure if it is so. Made a bet, waiting for the pit to form :) Freezing and other signs tell about it, repeated time intervals. Anyway we have to wait for confirmation :) If let's say the pit will form in the other direction I will manage to jump out in time, without losses or even catch up. So far, only the_SP500 index has started to bounce, the market is frozen.

 
So far it's unclear what's going on, or should I say nothing at all:) _SP500 is still kicking, everything else is silent:)))) Something is about to break through:) The formation of a funnel is in doubt, the calm on the ranch has stabilised...

The slump of the pair was caused by some news, but it did not even closely resemble a vortex, I can tell the difference :) At the moment the relevance of the funnel appearance is over :) Let's wait:)
 
I think analysis of ticks, especially those poorly informative after DC filters, is not able to give me anything... I was working with them for a long time, even made a couple of Expert Advisors, analysed speed of ticks, speed of price changes, caught shapes, etc., but it was useless. All I have left is not the best Pips algorithm, which I do not even use now. And this is the maximum what can be squeezed out of ticks (IMHO), and it's not easy, because even the tester is not a helper here. For more it is necessary to analyse a huge amount of ticks and there is not much sense comparing with standard TF charts.
 
Piligrimm, can you tell me how you can use MT to chart a close of any currency pair on a chart of another currency pair?
 
2Figar0 I agree about filters, there are filters that filter out strong drops, squeezing out such intermediate ticks, managed to identify such filters in the same comparison of two DTs only taking into account time output when stretching ticks. It was clearly seen why there were three ticks up and down 6 pips high, when in another DC they were 1 pips apart. Apparently the sensitivity of the filters is set differently. Now I can say that the analysis between two DTs, if it makes sense, only to detect this kind of filters. As you can see there are not ten differences, but only two, but what do they give us, only more complete information about the imbalance as such.
 

This is not due to overloading of the broker server, faster arrival of changed quotes, but only the difference in filter settings, everyone adjusts as they see fit. Yes, that is why thick-skinned EAs are a necessity, because one brokerage company will do one thing and another one will do another. The filter works in such a way that it selects the most realistic quotes, taking into account timeframes, but it does not emulate quotes, but only rejects some variants, leaving only the most plausible ones. Unfortunately, this is not the truth they want to see for the user or the program, that's why such topics as analysis between brokerage companies begin.

Note that I use two indicators at once, server and client time intervals, so there is almost no difference in the charts, only in these filtered areas on the server side.

The time difference, is calculated in eight-byte time, where the server date is converted to (gcnew DateTime( 1970, 1, 1 ))->AddSeconds( iSrv ), then the sum of the tick time difference, server and client, divided by nine to the eighth power is used, in this graph, to get the difference in seconds, you must divide by ten to the seventh power. This way I can infer with high accuracy excluding problems in data update rate. Except that one pixel per tick is consumed, but I think for some modes, like outputting time within ticks, I'll remove the consumption, then it will be perfectly comparable even in size. What can you do, I'm a digger, even without wanting to dig to the root:)

Dear Pilgrim, I am interested in what you have to say in response to this statement?

1234567891011121314
New comment
Reason: