Machine learning in trading: theory, models, practice and algo-trading - page 2334
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
The question does not specify to what to apply the luck and systematicity. If to the TC, it depends on the TC, if to the external conditions, then the rare cases are initially not systemic. And may be exceptions to the rules. Eurodollar 14 from May to March '15. Not a systemic case.
Talk about representative TC statistics. In general, let's move on.
If the average position holding time for the TS is 10 minutes. And the current position is frozen for 10 hours, then its result is a random (completely non-systematic)?
If the current position is the result of a system operation, then it should be taken into account. The question arises - what does " the current position has been frozen for 10 hours" mean?)
. Eurodollar 14 from May to March '15. Not a systemic event
All "cases" that happened in the market are systemic. That's the right way to look at it.
If the average position holding time for the TS is 10 minutes. And the current position hangs for 10 hours, then its result is a random (completely non-systematic)?
Absolutely unambiguous answer is impossible in principle - only relative to the chosen probability model.
For example, consider time to be normally distributed:
1) We estimate from a large sample the correctness of such an assumption in principle.
2) Using an average sample we estimate parameters of a normal distribution - the average is not enough, we also need the variance and the sample size
3) Using the smallest (most recent) sample we estimate whether there was a substantial deviation from the calculated parameters.
If the average position holding time for the TS is 10 minutes. And the current position hangs for 10 hours, then its result is a random (completely non-systemic)?
If the logic of the system's trade opening is tied to its own market time - that's fine. Just to see if the "binding" was correct in this "case". Just in case. Well, if it is not bound... No one can say anything useful here, because anything can happen.
An absolute unambiguous answer is in principle impossible - only with respect to the chosen probability model.
For example, consider time to be normally distributed:
1) Using a large sample, we estimate the correctness of this assumption in principle.
2) Using an average sample we estimate parameters of a normal distribution - the average is not enough, we also need the variance and the sample size
3) By the smallest (most recent) sample we estimate whether there was a substantial deviation from the calculated parameters.
It would be good to have an example where 10 hours with an average of 10 minutes is systematic.
A good example would be when 10 hours with a mean of 10 minutes is systematic.
In a purely theoretical sense, this could be a thick-tailed distribution (Cauchy, for example) that has no expectation, which makes the arithmetic mean of the sample not much sense.
In a practical sense, the obvious example would be a trend change by a channel (for a trend system). In terms of theory, this means different distributions in the sample, which also reduces the value of the sample mean as a kind of benchmark of correctness.
PS. For a random walk, it seems that the expectation of time to reach the level is just not defined (so the arithmetic average of the sample does not add up to anything).
A good example of 10 hours with an average of 10 minutes would be systemic.
10 hours averaged 1 time per year. 10 minutes averaged N times an hour.
A good example would be 10 hours with an average of 10 minutes is systemic.
A holiday of any kind. The market can freeze for days.
A holiday of any kind. The market can freeze for a few days.
Well, we know what we were talking about from the beginning. I didn't make any reservations for formalism's sake.