From theory to practice. Part 2 - page 104
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 remember hunting tarantulas as a child. A ball of plasticine on a string in a burrow and you pull it out. Big, black, furry and very poisonous. The breadth of our Voronezh.
Personally for me the use of SB is interesting only as a criterion to check the results of a prediction. For example, I try to predict the direction of bar closing of the day with the help of some Shamnsko-Koldun indicator. If the result is plus or minus 50/50, then I do not use these magic tools. And if the result is 60/40, or 65/35, and moreover on the big statistics, and evenly over the years - you should look at them carefully.
This can be formalized as a test of the null hypothesis that the binomial distribution parameter is equal to a given value relative to the alternative hypothesis that the parameter exceeds this value. The hypergeometric distribution, on the other hand, can be used to test the hypothesis of uniformity in years. This approach is simply more convenient for me (when everything is considered more or less strictly), although I am well aware of its limitations. Matstat is just a tool, not a way to describe the essence of regularities.
The impossibility of making money on SB is an obvious fact. There was a question here in the thread - what is the point for traders in this fact? For me (and many others) there is a point and it is very simple - it is necessary to look for price differences from SB. This approach is formalized by means of methods of hypothesis test theory from matstat. What I wrote earlier about zigzag is one possible example of this very standard approach. I have another example in my article about gaps.
Yes, it's a classic, in professional circles. But scammers and traders do not like (do not want) to do that, as well as forward testing, though exactly they do it not once but hundreds of thousands of times to optimize forward)))). And in any, even the most run-down prop, not to mention the noble hedge funders, forecast modules are constantly tested on noise (sterile) data, God forbid that the "SB" acuracy of some MO model was more than 50.01% and the correlation predicted / real retourne above 0.001(of course figures may vary depending on frequency of data), in general "SB" - nada should be on all 100500 statistics and smart econometric models from young students of red diplomas, which rarely survive longer than half a year in such organizations where you need alpha and not science-y crap from recent articles.
It is not necessary at all, in fact it is probably the most difficult way to find patterns.
If the strategy gives a plus in SB, it means that somewhere there is a mistake.
You are ahead of the curve. All right, but in reality do not use straight to the SB, and rather specifically generated data, which for a number of statistics imitate real data, which are usually very large (tens of gigabytes per day) and there more than 2 / 3 are not price series and generally the date is not with exchanges (orders, transactions, etc.), it is a separate fairly large-scale work, but the essence of the same "on the SB should not be plus.
This can be formalized as a test of the null hypothesis that the binomial distribution parameter is equal to a given value relative to the alternative hypothesis that the parameter exceeds this value. The hypergeometric distribution, on the other hand, can be used to test the hypothesis of uniformity in years. This approach is simply more convenient for me (when everything is considered more or less strictly), although I am well aware of its limitations. Matstat is just a tool, not at all a way to describe the essence of regularities.
After reading this post, I was reminded of Vladimir Vinokur's monologue - The Great and Powerful Russian Language ) )))
So lucidly painted the rule of making a profit with SB.))
Not to mention the noble hedge funds, the forecast modules are constantly tested on noisy (sterile) data
Where's the flavor coming from? Been working there?
You're digging in the wrong place...
I can say with almost 100% probability that almost all indicators will shoot a signal reversal almost in the middle of the trend.
If someone has an indicator that gives mirror signals on real and demo trades, it's just what the doctor ordered.
And about SB...
The only question here is - what are you smoking? ;)))
Where did the tobacco come from? Have you worked there?
Yes, I used to work there, in principle I still do, but I don't go to the office like a pioneer every day any more, I work remotely, it is of course a different type of work and the money is much less, but as before I cannot sleep for three days and burn to the full extent of my health.
Yes, it's a classic, in professional circles. Scammers and marketeers, on the other hand, don't like (don't want) to do it,
Banging on the wall next to an open door? :))))) What's the point?
Banging on the wall next to an open door? :))))) What for?
Not everyone has a magic door to Narnia) Tell us how it is, if you're lucky)