For those who have (are) seriously engaged in co-movement analysis of financial instruments (> 2) - page 21
![MQL5 - Language of trade strategies built-in the MetaTrader 5 client terminal](https://c.mql5.com/i/registerlandings/logo-2.png)
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 wonder if this tool can be used in a cluster?
In all variants (of which there are two main poles: trend (trendomat, Yuri's portfolios) and contrend (arbomat, recycle))
implicitly some kind of predictions are used.
This is not a forecast, but a deliberate adjustment to history in both cases.
The simplest example of when the r-portfolio, calculated for stocks included in the DJI index, failed to predict the fall and went down: "During Thursday's trading, the Dow Jones fell below 10,000 points - by more than 9%" - http://www.gazeta.ru/news/lenta/2010/05/07/n_1491879.shtml.
It's not hard to see the history adjustment on Recycle either, because it has channel properties only on Sample and loses them on OOS
In other words, it is naive, to say the least, to rely on the portfolios compiled by either of the above two methods as a forecast.
I also understand that you can be interested in different ways, some just theoretically, some practically. If you want to estimate the results, we should analyze them (for example: such-and-such pairs with such-and-such coefficients make a channel with such-and-such durability (it's not obligatory)).
+1
theoretical construction methods are useless to discuss. They cannot be better or worse in principle. Without practice - a trading idea formalised in code - there is nothing to discuss. Then you can analyse, apply other methods and compare them.
Cool topic, I read it in the same breath, already testing the Correlations script - in an orgasmic fit. Mathematicians burn it up again! But this is a question, does anyone have a history of major financial instruments for ten years, at least in the form of daily bars. I'm sure it all can be found, but I don't know, if you don't mind, please tell me. First of all I am interested in fundamental instruments: Index (SP, DJ, FTSE etc.), CRB-Index, T-Bond, Notes, Gold, Dollar Index, WTI etc.
Still, I would be very wary of intraday timeframes. You could apply linear regressionanalysis and find out that the overnight low-volume flat of many instruments is very similar to each other, and start day trading on that basis.
Cool topic, I read it in the same breath, already testing the Correlations script - in an orgasmic fit. Mathematicians burn! But this is a question, does anyone have a history of major financial instruments for ten years, at least in the form of daily bars. I'm sure it all can be found, but I don't know, if you don't mind, please tell me. First of all I am interested in fundamental instruments: Index (SP, DJ, FTSE etc.), CRB-Index, T-Bond, Notes, Gold, Dollar Index, WTI etc.
Still, I would be very wary of intraday timeframes. You can apply linear regression analysis and find out that the overnight low-volume flat of many instruments is very similar to each other, and start day trading on that basis.
Yes indeed! Multicurrency analysis works wonders. My results are amazing on medium and long term as well.
Almost all instruments with very deep history can be downloaded for 10 rubles apiece here. You need to register.
Cool topic, I read it in the same breath, I'm already testing the Correlations script - in an orgasmic fit. Mathematicians, keep on burning! But this is a question, does anyone have a history of major financial instruments for ten years, at least in the form of daily bars. I'm sure it all can be found, but I don't know, if you don't mind, please tell me. First of all I am interested in fundamental instruments: Index (SP, DJ, FTSE etc.), CRB-Index, T-Bond, Notes, Gold, Dollar Index, WTI etc.
Hardly available in the public domain. This is commercial data. You can buy it at appropriate stock exchanges or from quotation providers: Tenfore, Reuters, IQFeed, ESignal ...
Some of them can be downloaded from finam free of charge, but the quality of quotations is not super.
Yes. Indeed! Multicurrency analysis works wonders. I have also had great results on the medium and long term.
I'm sure it's all available somewhere, but I don't know where, if it's not hard to find, please tell me. I am sure that it is possible to get it all somewhere, but I do not know where, if it is not difficult to prompt please. First of all I am interested in fundamental instruments: Index (SP, DJ, FTSE etc.), CRB-Index, T-Bond, Notes, Gold, Dollar Index, WTI etc.
Cool topic, I read it in the same breath, I'm already testing the Correlations script - in an orgasmic fit. Mathematicians burn it up again! But this is a question, does anyone have a history of major financial instruments for the last ten years, at least in the form of daily bars. I'm sure it all can be found, but I don't know, if you don't mind, please tell me. First of all I am interested in fundamental instruments: Index (SP, DJ, FTSE etc.), CRB-Index, T-Bond, Notes, Gold, Dollar Index, WTI etc.
Still, I would be very wary of intraday timeframes. You could apply linear regression analysis and find out that the overnight low-volume flat of many instruments is very similar to each other, and start day trading on that basis.
Try it here on Finam:
http://www.finam.ru/analysis/export/default.asp
They focus on arbitrage and synthetics, and forget about other opportunities, such as cluster analysis.
Semenych has published a couple of articles, but it is not easy, because what he proposes to do (not analyse, but do!) is not necessarily profitable.