Random Flow Theory and FOREX - page 77
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 have never given signals to anyone. _
I assure you it wasn't me. I never gave signals to anyone. I used to participate in some contests on Broko (but it was a long time ago). I do not have a website and was not. I am sorry somebody is using my nickname, I hope it is not for evil. If you suddenly stumble across this information, please drop me a link in a private person, if you do not mind. Thanks in advance.
Better also told me he saw my signals, but that can't be.
Is pfsignal.com your website? Somewhere I found a profile of some Prival with a link to this site.
What's in the way?
What else is there to stop the dancers? Only the lack of a ticking history and the presence of a spread.
Otherwise, beautiful marquise, all is well, all is well! (c) A song of some kind.
The article discusses the Kalman filter as such, which has the widest application.
in your sound -- almost a squid.
What else is there to stop the dancers? Only the lack of a ticking history and the presence of a spread.
Otherwise, beautiful marquise, all is well, all is well! (c) A song of some kind.
What is the strategy in a nutshell? I used to follow this thread, but I forgot. The only thing I remember is Kalman's filter and the tick story.
You are wrong. I will try to write briefly about what it is.
It is not a question of my being right or yours.
We have absolutely different approach. I am not interested in algorithm of filter calculation. I am interested in using this filter, which is not a simple task at all. Perhaps you will need to know its calculation algorithm when using the filter, but agree that in this case there is no discussion of the calculation algorithm itself. Moreover if I take ready-made code, e.g. EViews, over 20 years of its use thousands of smart guys fished out all bugs and removed all questionable points I can just trust. At any rate, when using off-the-shelf code I can get results that don't contradict similar ones.
1. You are right that the procedure of calculating the Kalman filter is known and has been known for a long time. The order of calculation and formulas themselves are given here in a branch (I even laid out two ways, depending on a priori data how to count these matrices), but...pay attention 4 years as lay, and not a single implementation is not laid out in code base....just ask yourself why ? (although I could be wrong it's been a long time since I looked there) but what they sometimes send me on Skype can hardly be called a competent solution...
2. in order to use this filtering procedure correctly, you need to understand it, really understand how it works, otherwise you get nonsense.
With my approach, there's no problem using the Kalman filter per se. There is the problem of using the model within which the Kalman filter is used. This is a very widely used state-space model in economics, in which the output depends not only on the input, but also on some state of the model. Yes the matrix problem remains, but it becomes meaningful because it is part of the model, which is the core of the TS.
.
Once again. "Everything before us has been stolen" and God forbid we should be able to use what we already have. The packages I have named are: EViews and R. The composition of the R package has been laid out. Kodobobase has a wrapper for R. Free package. The de facto standard in the application of statistics in economics.
I started posts on econometrics a year ago, hoping to grow a MQL-like hangout. You are one of the potential candidates and it is with great regret that I look at your attempts to answer questions for which the answer has long been known.
That's your logic, faa.
You think people who haven't mastered multiplication should start counting integrals straight away.
Mastering doesn't start with downloading EViews.
And neither do other packages.
Mastering starts with a simple model, with simple examples, with doing laboratory work.
In the case of forex, you cannot use programmes such as e-views, because they will not help you at all.
Because the essence of calculations and data preparation is quite different from what you can do
in all these programs.
And the calculation of the trading model is not there either.
That's your logic, faa.
According to you, people who haven't mastered multiplication should immediately start counting integrals.
Mastering doesn't start with downloading EViews.
And neither do other packages.
Mastering starts with a simple model, with simple examples, with doing laboratory work.
In the case of forex, you cannot use programmes such as e-views, because they will not help you at all.
Because the essence of calculations and data preparation is completely different than you can do
in all these programs.
I will allow myself questions on the subject, as I understand it, closer to you.
1. Is it possible to program in MQL without knowing assembler?
2. Can you program in MQL, without having written any compiler?
Or maybe the problem of using MQL has nothing to do with assembler and the experience of building compilers? I think the answer is obvious. The problem is WHAT to program, and using MQL is a matter of technique.
It's the same here.
We write TCs, which in other terms could be called models. The problem is in the models. And quite a large number of them have been developed. There are ready libraries of programs for models, I have named two of them, but there are much more. You have to know how to use them. It is one thing to use indicators from kodobase, and it is another thing to use the same indicators but understanding that they are regressions, most of them are autoregressions. And to use off-the-shelf code for implementation. There used to be a branch here on MNCs (like this one about Kalman). There was a discussion about the intricacies of implementation. Any stat package starts with an ISC. No need to discuss and stress about anything. It's just another level, as you can apply ISC and at the same time see and try a bunch of other methods of estimating model parameters that you may not have heard of before. This is the beauty and benefit of using off-the-shelf packages.
Of course, there's a lot to learn before using regression analysis. That's for sure. But there are a fair number of people on this forum who have that initial training. Prival is among those people. He can take the next step in the form of mastering a specialist package. There are 700 people who have downloaded the R wrapper, which you just can't try as an indicator. So just for the sake of curiosity they will not download the wrapper.
My posts are addressed to these people.
Once again. "Everything before us has been stolen" and God willing, we should be able to use what we already have. The packages I named: EViews and R...
What is the annual pips yield in euros for example (fixed lot, with accounted spread)?