The perfect filter - page 2

New comment
 
gunia: P.S I just want to understand what you mean.

Judging by the vague manner of presentation and the negativity towards statistics, the author of this post is not too familiar with matstatistics:) Perhaps all the author's knowledge of statistics is limited to the fact that the "classic of the genre" is an experiment with a coin.

In any case, someone who knows at least the basics of matstatistics would express himself much more concretely.

 
J.B:

I don't know if I understand you or not, but in my opinion the best way to ignore small movements is as follows:

- There is a trend from point 5.0 to 5.2;

- Convert the amount of movement into a percentage: (5.2 - 5) / 5 = 4%

- We compare it with the set threshold, for example 10%; if we understand that the trend is small, we skip it.

- If necessary, we also cut off a large trend.

- In my opinion, it is better not to get addicted to different indices.

 
Mathemat:

Judging by the vague manner of presentation and the negativity towards statistics, the author of this post is not too familiar with matstatistics:) Perhaps all the author's knowledge of statistics is limited to the fact that the "classic of the genre" is an experiment with a coin.

In any case, a person who knows at least the basics of matstatistics would be much more specific.

I understand that the status of a moderator with such a sonorous nickname (which already hints at involvement) obligates to make a weighty statement in the topic about statistics... But it's a bad manner, to evaluate other people's knowledge or ignorance of something, if you don't have the patience to read the post carefully or are not all right with the ability to comprehend what you've read... My advice to you - if my post you have seen a negative attitude to statistics and "vagueness", try for the future not to save on snacks (and health in general, it will also be useful). P.S. Just in case I will add that I for a long time already do not play in a sandbox (in sense I have grown from kindergarten age), therefore I am not fallen for offers such as "and let's measure with whom...", and here I do not intend to engage in bullshit, proving better someone or not I am familiar with basics of statistics.
 
gunia:

Agree and disagree at the same time, it's just not clear what you mean.

What kind of data is not enough? How much would be sufficient? Price, volume, Level2, etc.

What is "reliable" data?

Please provide us with a concrete statement from a "serious" work, which proves that there are limits to the sample of data for statistical analysis. Especially interesting about "reliability".

P.S I just want to understand what you mean.

I don't want to pretend to someone else's laurels, so if Google can't help you in search of theses from "serious" works, then Mathemat is your help! :) And now a little on the subject: what market information do we have for research? The answer to this question can be obtained by running in the Strategy Tester an Expert Advisor on the timeframe that interests you. How many bars we will obtain in the history with the modeling quality of at least 90%? This is what I mean by conditionally "reliable" data for analysis. And this data is not enough for statistical studies. If you want, we can continue to communicate in private, because (in short, see above)
 
Wangelys:
The answer to this question can be obtained by running in the tester an Expert Advisor on the timeframe that interests you. How many bars are obtained in the history with the quality of modelling at least 90%? This is what I meant by conditionally "reliable" data for analysis. And this data is categorically insufficient for statistical studies.

Now I roughly understand what you mean by "sufficiency" and "reliability".

I use different tools for statistical analysis, not only MT's tester, all of which I find interesting and have enough time to master them.

I perceive MT more as a trading platform than as an analytical one.

 
Wangelys: To be on the safe side, I should add that I haven't played in a sandbox for a long time (I mean I have grown out of kindergarten age), so I don't fall for suggestions like "let's measure up to who...", and I'm not going to bullshit here, proving I know the basics of statistics better than someone or not.

Thank you, but your response has already convinced me that I have nothing to measure with you.

In the future, try not to speak out on issues in which you understand nothing at all (statistics, for example). It will save you nerves and energy.

You had a chance to make your point (=statistics) in your response to me, but you didn't bother to do so either.

If you think you do understand something about it (and you feel you have to prove it to others), then show it - and I will publicly admit that I was wrong to accuse you of the opposite.

 
Mathemat:

Thank you, but your answer has already convinced me that I have no argument with you.

Poor thing, I truly sympathize with you... That's probably why, like so many other lamenesses, you have such psychological complexes... A mixture of boorishness and megalomania. Who are you to take on the role of evaluating my (or anyone else's) knowledge? Who are you to "give me a chance" to convince you of anything in: "You had a chance to make your point (=statistics) in your response to me"? (Besides the word answer implies that there was originally a question, while you didn't ask me anything, you just rudely announced your presence in the forum thread) .... For me you are nobody, and even nobody with a small letter, and something to prove to you...(read parables in the Bible about uselessness of such an occupation). And on this part of your statement: "...I publicly admit that I was wrong when I accused you..."I want to put a little emphasis - I think that a public confession of your wrongness would be very appropriate, because you clearly had no moral right to accuse me - no one gave you such authority. You admit your wrongness and apologize to me, and I will apologize for my harsh reaction to your behavior ...(if you wish).
 

Thank you gentlemen for your help.

To tell the truth I didn't expect such a simple task to turn out to be quite simple, not even technically but conceptually.

I've tried a lot of indices, at first I've stopped at fast double SMA from minuets and count the difference in vicinity, with a certain step, adding in the loop those places that were in a certain range. Then I tried to do a zigzag. I've tried a lot of other things. In general, I failed so far.

But thanks anyway, thanks for all your help!

If I ever get something useful, I'll post the results, if anyone is interested.

 
Wangelys: Poor thing, I sincerely sympathize with you... That's probably why, like so many other flawed people, you have such psychological complexes... A mixture of boorishness and delusions of grandeur.

I don't care what you think of me. You have complexes, not me. I just appreciated your extremely vague and inadequate advice about statistics.

Keep your emotions to yourself, we're not girls here.

 
Wangelys:
Keep your emotions to yourself
12345678910
New comment