The St Petersburg phenomenon. The paradoxes of probability theory. - page 15
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
Hurst is not mine, I am closer to Shepherd with his H-volatility.
Not the point, what matters is the presence/absence of an analytical view. Another (more difficult to formalize) question is the correspondence of statistics to some class of TS. How, at least approximately, can one trade when there is a noticeable deviation of Hearst (or your H-volatility) from their typical SB values?
So maybe you really need a screwdriver rather than tweezers? I generally prefer a hammer, as it knocks all sorts of things out of matter very quickly. I can't get it out with tweezers.)
What do you need? You tell me, you tell me, what do you need, what do you need, I can give you, I can give you, what do you want...
What do you want? You tell me, you tell me, what do you want, what do you want, I can give you, I can give you, what do you want...
Me? I don't want anything. Why are you bothering me? Maybe you want to tell me what you want.
Me? I don't need anything. Why are you bothering me? Maybe you want to tell me something.
I'm sorry. Tactical move. Waiting for your opponents, if there are any. Don't sleep for half an hour, please.
Sorry. Tactical move. Waiting for your opponents, if there are any. Stay awake for half an hour, please.
Yeah, I'll clean the hammer, grease it.
Not the point, what matters is the presence/absence of an analytical view. Another (more difficult to formalise) question is the relevance of statistics to some class of TS. How, at least approximately, can one trade when there is a noticeable deviation of Hearst (or your H-volatility) from their typical SB values?
Pastukhov gives the answer: look at the history, consider which pair is trend-flat. Major pairs are trending, H>2; Minors are flat, H<2, H=2-SB. We trade the trend ones along the trend, the flat ones against it, but the advantage is minimal, direct dependence on the spread, the spread should be minimally small, or even better, without it, to make this strategy work. The main thing: at rather long trade the IR is practically zero, though in some places it can be even more than zero. That's the paradox, you know, but you won't take it.
Pastukhov gives the answer: look at the history, consider which pair is trend-flat. Majors-all trending, H>2; minors-flat, H<2, H=2-SB. We trade the trend ones along the trend, the flat ones against it, but the advantage is minimal, direct dependence on the spread, the spread should be minimally small, or even better, without it, to make this strategy work. The main thing: in long enough trade the IR is practically zero, though in some places it can be even more than zero. Here's the paradox, you know, but you won't take it.
I don't really get it, but I suggest building a Monte Carlo distribution of H for a random walk and see what its quantile corresponds to H=2.
Pastukhov gives the answer: look at the history, consider which pair is trend-flat. Majors-all trending, H>2; minors-flat, H<2, H=2-SB. We trade the trend ones along the trend, the flat ones against it, but the advantage is minimal, direct dependence on the spread, the spread should be minimally small, or even better, without it, to make this strategy work. The main thing: at rather long trade the IR is practically zero, though in some places it can be even more than zero. That's the paradox, you know, but you can't take it.
Well, Pastukhov said...
Me? I don't need anything. Why are you bothering me? Maybe you should say something.
That's it. I'm sorry. Put the hammer away.
We construct some statistics on the price series. Using the criterion of agreement we check how much its distribution differs from what it would be if prices were a random walk. If the difference is statistically significant, it can indicate the possibility of trade. Of the criteria of agreement Kolmogorov-Smirnov seems to be the most appropriate.
Also, this criterion (and many others) would be very useful in the "From theory to practice" thread).
Tests of"how much its distribution differs from what it would be if the prices were a random walk" are of no particular value or utility.
Even the phrasing itself is wrong:"If the difference is statistically significant, it may indicate a possibility to trade". That is, otherwise trading is impossible, according to you.
This is a deep delusion. You have accepted a false premise as an axiom without even trying to verify it.
Think about the fact that the trading process is external to the price series being traded.
The statistics of the trading process are not reducible to the statistics of the price series at which the trade is made.
Conduct an experiment:
1. Generate a SB process.
2. Apply the trading rules to this SB process.
3. Ensure that it is possible to trade successfully on this SB process.
4. Repeat steps 1,2,3 many times, recording the results of the experiments.
5. Confirm the fallacy of the postulate that it is impossible to trade successfully on the SB process.
6. Determine the statistics of the trading process.
7. Compare the statistics of the trading process with the statistics of the SB process.
8. Finally, draw conclusions.
If you decide to do such an experiment, its results, if you present them here, will help you and many others to open their eyes and get rid of the closed-mindedness artificially created and thoughtlessly accepted, like the ubiquitous theory-delusion about "market efficiency". I hope this "market efficiency" fallacy doesn't mislead you.
SO
What program you do it in (R or not) is a moot point.