Adaptive digital filters - page 11
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 didn't express myself a bit accurately. If returns is a discrete series of +-1 pips, then mine is more accurate, the output of the filter gives an estimate, i.e. results in fractions of a pip.
So you managed to describe the input signal so accurately that it almost matched the original signal?
I didn't express it accurately. If returns is a discrete series of +-1 pips, then mine is more accurate, the output of the filter gives an estimate, i.e. the result is in fractions of a pip.
So, you managed to describe the input signal so precisely that it is almost the same as the original one?
If you knew the original :-), then you'd have something to say. It is a question of checking adequacy. And I don't know how to check it. Let's take two models. The first is say a Wiener process and the second is my model. How to know which model is better (more appropriate). I have already tried the mathematician, he says I do not know. The North Wind may give you some advice. I should at least read a book on adequacy testing, preferably with pictures (to be clearer). You said something about drift, maybe there is a test
OK, nonsense. If you want, you can also use some numbers before that, i.e. price history :)
Just calling the quote a broker gives the "truth" (the benchmark, the ideal) is wrong.
The correct benchmark is the real ratio of the real exchange rate of one currency to the real rate of another currency.
Not with rounding to 4 decimal places, filtering, artificial change and other "gimmicks", but the real price.
And one should not try to predict the next broker's feat (changing the filter, going for stops, etc.), but the change in the real exchange rate.
And the broker, no matter how you look at it, will come to him. Sooner or later, but it will come.
Well, yes, Andrei, you are right. But no one knows the real exchange rate, except maybe Bernanke (and it's unlikely). I do not want to get into the maze of FA, and so all I have are the quotes of this brokerage company. That's what we do, we predict the antics of our brokerage company. Isn't it so?
Well, of course he will, where will he go? Only this information is interesting for the investor, not for the speculator.
That's what we do, we predict the antics of our brokerage companies. Isn't it so?
Of course he will come, of course he will. Only this information is of interest to the investor, not to the speculator.
Yes, I got a little carried away here. There is a notion of a posteriori probability. I wanted to emphasize that all the same quotes are not accidental, especially - past ones, but I formulated it incorrectly.
Well, I am not pretending, just for the purity of my thoughts.
And one should not try to predict the next broker's deed (change of the filter, the stop loss campaign, etc.), but the change of the real rate.
And the broker, no matter how you look at it, will come to it. Sooner or later, but it will come.
Stop telling everyone my secrets! :)
Well, yes, Andrew, you're right. But no one knows the real exchange rate, except maybe Bernanke (unlikely). My trading robot does not want to get into the maze of FA, and that's why all I have are quotes of this broker. That's what we do, we predict the antics of our brokerage company... Isn't it so?
Well, of course he will, where will he go? Only this information is interesting for the investor, not for the speculator.
It is important where the price goes. Practically, you can win with simple strategies, yes practically by playing orgy and entering the market on a coin.
I was a little wrong in my expression. If it's a discrete series of +-1 pips, then it's more accurate, the output of the filter gives an estimate, i.e. the result is in fractions of a pip.
That is, you managed to describe the input signal so precisely that it nearly coincided with the initial one?
If you knew the source :-), you could say something. It's a question of checking the adequacy. And how to check it, I don't know. Suppose we take two models. The first one is say a Wiener process. The second one is my model. How to know whose model is better (more adequate). I have already tried the mathematician, he says he does not know. The North Wind may give you some advice. At least read a book on adequacy testing, preferably with pictures (to be clearer). You said something about demolition, maybe there is a test.
Mathematics has no direct answer to this question. At least I haven't found it in this formulation. It does not mean that others cannot try and may be more lucky. There are of course methods of evaluating adequacy, but they are not suitable for prices. The only criterion is regular profit.
Prival, here's another example in pictures. Look at the quotes from one of the DCs:
It is true that the picture is "polluted" by graphics, but still you can see how ugly it is. And here are the quotes of another (respectable) brokerage company about the same period:
The right side of the charts is about the same time. Now tell me: what confidence intervals we can talk about with such a difference in quotes? By what criterion are you going to choose "decent" DCs?
P.S. It is possible that there is a time shift, which I have not considered, but it is obvious that the top picture is a complete bred safe cable.