Discussion of article "Random Walk and the Trend Indicator" - page 5

New comment
 

вот про это я и говорю выкенте учебники эконометрики, если они  проверяют "значимость линейной регрессии", можно проверить значимость только коэффициентов в уравнении.

N.S. Kremer. Probability theory and mathematical statistics. Paragraph 13.3. It is called: "Testing the Significance of the Regression Equation". Probably this book should be thrown out too, even though it is not econometric, and the author should be pointed out that only the coefficient that exists by itself can be believed?
.

The argument is about nothing. I'm not interested. Sorry, I will try not to answer any more.

There is nothing to apologise for.
 
vlad123:

The topic is very interesting, there are questions

1. does the (short-term) market have a memory?

2. How are the values of the trendiness indicator obtained - there are some values there, how are they obtained? I didn't see the formula.

3. When you write about trendiness - does the new forex tick depend on the previous one? Or two ticks or how many more? But in general it can be written in. And in general it makes no sense to talk about trendiness, saying that your coin is non-trending, but it is true.

4. But from point 3. we can draw a very important conclusion

- you can try to find a trend if you remove the trendlessness from the real picture.

i.e. z(x)=x(x)-y(x).

Does the (short-term) market have a memory? There is no general answer to this question in the article. There is only this answer - if the market is trending or anti-trending, it has memory. If it is non-trending, then maybe not.

How are the values of the trendiness indicator obtained? I quote the article "We take the value of "++" + "--" - "-+" - "+-" as an indicator".

new for forex tick depends on the previous tick? If there is trendiness, it depends. When there is trendiness, the indicator shows it.

 
Virty:

Does the (short-term) market have a memory? There is no general answer to this question in the article. There is only such an answer - if the market is trendy or anti-trendy, it has memory. If it is non-trending, then maybe not.

How are the values of the trendiness indicator obtained? I quote the article "We take the value "++" + "--" - "-+" - "+-" as an indicator".

new for forex tick depends on the previous tick? If there is trendiness, it depends. When there is trendiness, the indicator shows it.

1. If the market has no memory, this article should be banned here as scaring away gamblers (the main consumers of the product).

2. On the indicator - suggestion - add the length of the movement ie example +7pips = "+++++++".

3.=1.

 
vlad123:

1. If the market has no memory, then this article should be banned here as scaring away gamblers (the main consumers of the product).

2. On the indicator - suggestion - add the length of the movement, i.e. example +7pips = "+++++++".

3.=1.

1. If the market has no memory - then this article should be banned here as scaring away gamblers (the main consumers of the product).

The article is devoted to the research "Does the market have memory". The trendiness indicator is specifically made for this purpose. I've bled myself to death to convince that the indicator shows trendiness (read "memory") over long periods of time.

2. On the indicator - suggestion - add the length of the movement i.e. example +7pips = "+++++++".

The article discusses measuring trendiness on long chains. I can repeat it. To measure by long chains you need a lot of members in a row (from 30, or better 300 or 1000). And who would need an indicator with a lag of 1000 bars?

1=3

 
There are pearls in this sea. Thank you for the article.
 

Save

 

Thank you very much to the author, I look forward to your new articles.

 
I suggest the author to write an article on the topic of The Discerning Bride. On one forum many years ago attempts were made to build a trading system on the basis of this problem.
 
Rosh:

I suggest the author to write an article on the topic of The Discerning Bride. On one forum many years ago attempts were made to build a trading system on the basis of this problem.
I did not understand the idea of the system. The bride has all grooms knowingly different. The trading system has all future trades the same.
 

Задача о разборчивой невесте, или проблема остановки выбора может быть сформулирована следующим образом:[1]

A bride is looking for a groom (there is a single vacancy).

There is a known number of applicants - n.

The bride communicates with the applicants in a random order, with each of them not more than once.

About each current applicant it is known whether he is better or worse than any of the previous ones.


As a result of communication with the current applicant, the bride must either refuse him or accept his proposal.

If the proposal is accepted, the process stops.


The goal is to select the best suitor.

Much attention has been paid to this problem largely because optimal strategy has an interesting feature: if the number of candidates is large enough (on the order of a hundred), the optimal strategy will be to reject all the first n/e (where e=2{,}718\,281\ldots is the base of the natural logarithm) applicants and then choose the first one who is better than all the previous ones[2]. As n increases, the probability of selecting the best contender tends to 1/e, i.e. about 37%.


The bride, is an advisor

Grooms are currency pairs

Evaluation according to the criteria of the TS for maximum compliance with the given parameters.

In this case, we can significantly move the stable trading postulate about the number of system parameters upwards, because

With a small number of filters/indicators, given a large number of candidates, many of them will be on the same face.... and the comparison becomes meaningless.

With a huge number of filters, - there will be no signals at all.....

However, the possible number of indicators at such approach, imho, can go beyond a dozen and further....

..... with the choice of indicators as criteria is a separate topic....

When evaluating the first n/e it is necessary to buffer the evaluation result

Stop overshooting as soon as a candidate with a score higher than any of the first n/e is found.

The primary source assures about more than 50% probability of choosing the ideal groom by searching 37% of applicants.

http://w ww.mccme.ru/mmmf-lectures/books/books/book.25.pdf

Thus, in this case, the chances of the princess making a successful choice

(with optimal strategy) are greater than 50%.

123456
New comment