Analysis of multiple currency pairs by currency, your opinion, can this be used?

 

What do you think, is it possible to react to the flows with the help of average currency pair analysis, using, say, gold in comparison, what reactions in the market have you noticed in the flows, say, is it possible to identify the movement in one direction or another of a particular currency pair using the average data on the movement against other currency pairs, well as against gold, say, if we take all currencies and depending on their ratios, to draw conclusions. I wonder if it is worthwhile to do this or it is a futile idea to track the movements of relativity? I would like to hear all opinions.

Basically all currency pairs can be levelled at one point, give them an acceptable equivalent and catch the difference in direction instantly, for example fluctuations in one currency pair can say something for another currency pair or say a clear movement of a particular currency from the comparison set, etc.

Yes I read the article on cluster indicators after writing this post, but I am interested in opinions on whether this is worth working with as analysis for an EA.

P.S.: Is this nonsense or not?

 
xnsnet:

What do you think, is it possible to react to the flows with the help of average currency pair analysis, using, say, gold in comparison, what reactions in the market have you noticed in the flows, say, is it possible to identify the movement in one direction or another of a particular currency pair using the average data on the movement against other currency pairs, well as against gold, say, if we take all currencies and depending on their ratios, to draw conclusions. I wonder if it is worthwhile to do this or it is a futile idea to track the movements of relativity? I would like to hear all opinions.

Basically all currency pairs can be levelled at one point, give them an acceptable equivalent and catch the difference in direction instantly, for example fluctuations in one currency pair can say something for another currency pair or say a clear movement of a particular currency from the comparison set, etc.

Yes I read the article on cluster indicators after writing this post, but I am interested in opinions on whether this is worth working with as analysis for an EA.

P.S.: Is this nonsense or not?


'Theoretical basis for building cluster indicators for the FOREX market' This is where your point is made. It's called cluster indices. Sounds like a life-affirming idea. But haven't figured it out yet - waiting for a sleepless night.
 
Yes, I was just talking about this article, I forgot to give a link in a hurry, although I know how to use the search:))) I've been thinking about such an analysis for a long time, but now I got excited and wrote a post and finally got an article:) By the way, it is an interesting indicator, it has something what I was thinking about:) As they say, people do it for the people:)

For all that I slightly misunderstood the averaging of currencies, although it is indeed an indicator for visual recognition of a movement:) I think it is worth experimenting with it. I am sure there are some users of this indicator.

In fact, you need to know the departure of the currency, not its longitudinal flow, trace the departure till the identification and then reset, that is why what you can see in the indicator is not exactly what you can really use in an EA, and it is of little use, the same indicator mouwing averaging illustrates this approach, but the matter is even more profound:) In short, to display it on an indicator chart I cannot even imagine how, it's much easier to imagine such a variant of tracking and try to implement it in an Expert Advisor. Although, no, I suppose it is possible to draw spikes. The main thing is to think and the main thing is to come to the goal that I thought at the beginning and not to what I see, but what I see almost fits the description, it is a pity that almost.
 
 
xnsnet:

What do you think, is it possible to react to the flows by means of average currency pair analysis, using, say, gold in comparison, what reactions in the market have you noticed in the flows, say, is it possible to identify the movement in one direction or another of a particular currency pair by means of average data about the movement relative to other currency pairs, well as relative to gold, say, if we take all currencies and depending on their ratios, to draw conclusions. I wonder if it is worth it or if it is a futile idea to track the movements of relativity. I would like to hear all opinions.

Basically all currency pairs can be levelled at one point, give them an acceptable equivalent and catch the difference in direction instantly, for example fluctuations in one currency pair can say something for another currency pair or say a clear movement of a particular currency from the comparison set, etc.

Yes I read the article on cluster indicators after writing this post, but I am interested in opinions on whether this is worth working with as analysis for an EA.

P.S.: Is this nonsense or not?


Read Larry Williams, it describes a similar technique called Will-Spread.
 

Thanks for the links, all very interesting and informative! Looking into it:)

Z.U.: If the profit is only one tick, take it. Long-Term Secrets of Short-Term Trading.

Really liked the conclusion of the book, and the book itself:

Not only is this business not black and white, life is not black and white either. We all know this (I guess so), and yet as traders we want absolutes so badly that we absolutely forget to think. For example, mathematics is absolute, but when applied to the imperfect world of stocks and commodity futures, it becomes a tool simply to bring more clarity and certainty to imperfect concepts. Please never forget that speculation is above all a thinking business. If you are not very good at thinking, or at least at getting the right answers, I suggest that you look for where to jump off.

 
You are on the right track, group analysis of currency pairs is the future, gradually most traders will come to it. I have been using group analysis for a long time and the results of forecasts for a group of currencies are much better than when using one pair with lagged arguments. In addition, the use of lagged arguments greatly reduces the dynamic characteristics of expert systems, and in case of a group analysis using only the last formed bar - the system is able to react quickly even to news due to inertia of the market, and always be one step ahead of it. According to my observations the key to effective trend forecasting is taking into account the group dynamics of currencies, averaging reduces forecast accuracy, it is better to use straightforward values of signals, and for creation of systems based on pattern recognition use rationing of currency pairs by Bid, then all instruments will be on the same scale and the pattern will be unchangeable during moving from one instrument to another and at any market changes.
 

Thanks for your comment Pilgrim! By the way I have already encountered the fallacy of averaging.

I've read the book and thought why I haven't used stock indices, I haven't even considered it, at the moment I'm evaluating the situation in that direction as well.

I'm also thinking about collecting information by radio frequency, which is not very powerful in case of temporary communication blackouts or other collateral factors, but I'm interested in processing this kind of information in order to try things out. I will do it using my own library, by my Expert Advisor, storing actual fragment of history, by which imbalance can be estimated. That is all I have come to so far. In my opinion, it is much easier to play with information in my program, i.e. to carry out similar experiments and then try to implement them in my Expert Advisor. If this part proves to be more successful for investigation, then it seems to me that the EA will definitely not manage without the library, and especially not without other EAs, because we need timely processing of each tick from all currency pairs, plus saving into a file, and in one EA we have only one event of a new tick arrival for one currency pair. If anyone has any doubts about the necessity of such method, let me explain, it seems to me the key to successful analysis, bars are also logical averaging and therefore deceptive. In short, strange as it may seem, I must write a whole complex of research software for the sake of it, hopefully not too much, but I'm already glad that at least some concrete lead has appeared.

On the other hand breaks of communication are not so essential, anyway for the reason that currencies all the same should equalize in one point, and already from this point to look back, at usual certainly it is possible to miss something, but I think it not so essential, in the general category of conventionalities which much more, dump will not prevent.

P.S.: In head all looks beautiful, was formed as they say, it remains only to look at it in practice, if will turn out something interesting, I certainly will share!

 
xnsnet:

I'm also thinking of collecting information on the clock, although it's unlikely to have that much power, in case of temporary connection interruptions or other side factors, but for the sake of experimentation I'm interested in processing this kind of information. P.S.: Everything looks nice in my head, as they say, it only remains to look at it in practice, if something interesting comes out, I will certainly share it!


You can see the difference between frame construction and time construction here http://forum.fxclub.org/showthread.php?t=32942
 
As for the potic history - that's right, this approach will greatly improve the accuracy and dynamic performance of the models, all depends on computing resources, if your computer can handle such a calculation in real time - you will be ahead of the game. I also advise to consider as many instruments as possible, they are all related and have a strong influence on each other, the mutual dynamics between them - the key to success, and not only currencies, but also gold, futures, etc., even if you work only on the Forex market. If your computer cannot handle such volume, choose the optimal group of instruments, taking into account the ones you intend to trade, and the others in the group should have the highest correlation to them. In the case of a connection failure, if the depth of prediction of your expert system will exceed the time of the failure - it is not so bad, you will still have time to recover from the failure and make the right decision, if the depth of prediction is not enough - you have to apply additional measures, for example, make another forecast on M1 or M5, and use these predictions as a guide, and for setting the SL and TP for protection.
 
Piligrimm:
You are on the right track, group analysis of currency pairs is the future, gradually most traders will come to it. I have been using group analysis for a long time now, and the results of forecasts for a group of currencies are much better than when using one pair with lagging arguments.
I agree with you that multicurrency analysis can provide more information than a single pair. But it can be conducted in different ways. When I get to it tightly, I will do it in a different way, as described in the article 'Theoretical Foundations of Constructing Cluster Indicators for the FOREX Market'.
Multicurrency valuation techniques can be quite different, that's the point.
Reason: