Intuition testing - page 18

 
MetaDriver >> :

It took me a long time to figure it out and scratch my head... Got some of it, but not all of it, I'm afraid. :)

Yuri, could you write for future generations a universal RNG series generating series with most popular distributions for local mathematicians?

It can be done as an mql-object. That would make statistics fans happy! Make a topic, open a project, consultants and critics will come running in a jiffy, the topic will break the popularity record ... I'll probably be checking it out myself. But it's a little early for me to be in the lead on this topic. I'm not good at statistics. I'm catching up a bit, but so far so good.

You'll do just fine, it's your weight class. What do you think?

I don't see any problems with normalization to 1, but with distribution obtained this way it is too narrow and the more you sum up, the narrower it is.

If you want a wider bell, how do you do it?

After all, we are talking about specifying a distribution based on specified parameters. Suppose we set the RMS and RMS and the generator sets a set of random numbers that are normally distributed.

 
Urain >> :

Just normalizing to 1, I don't see any problems, but the distribution obtained this way is getting pretty narrow and the more you add up, the narrower it gets.

>> Well, yes, it is :)

Urain >> :

and if you want to get a wide bell then how?

Then it depends on whether you want to crack the problem in a general way, or whether you want to solve it in a limited range.

That is, for example, the way the problem was stated in the topic: find the probability distribution for 100 trials on the "intuition tester".

If a "limited" solution suits you, then it's not that complicated.

One can simply use Bernoulli's formula { Pn(m) = (n! / ( m! * (n - m!) ) ) * p^n * q^(n-m) } calculate the probabilities for each m (n=100, p=1/2, q=1/2).

For a rough estimate of the numbers double will suffice.

If you need "in general form" - then.... You should go to Jurix, Alexey-(Mathemat), and other Neutrons... :)

 
The latest issue of the Forex Store (#300) has an article on intuition in trading
 
MetaDriver писал(а) >>

It took me a long time to figure it out and scratch my head... Got some of it, but not all of it, I'm afraid. :)

Yuri, can you undertake and write for future generations some universal RNG, which generates series with most popular distributions for local mathematicians?

Now we can do it as an mql-object. That would make the statistics fans happy! Start a topic, open a project, consultants and critics will come running at once, and the topic will break the popularity record... I'll probably be checking it out myself. But it's a little early for me to be in the lead on this topic. I'm not good at statistics. I'm catching up, but so far.

You'll do just fine, it's your weight class. What do you think?

Thank you, of course, for the offer, but alas, I cannot because of the lack of time. I even work half-heartedly on my own system, because of various other things.

If you need to comment on something, no problem. The problem is really not complicated - I just need to convert the integral distribution function F(x).

And one more thing. IMHO, of course, but it seems to me that this script will not be popular at all. The statistics is well known to many traders, but its application to forex, in a constructive manner, with the possibility of obtaining a statistical advantage is a very difficult task. Few people have got to this point. Neither have I yet. And to do it for "bold experiments" of the masses, believe me, it makes no sense.

 

Test:

first time 46%

second time (one minute after the first time) 62%.

 
meta-trader2007 >> :

Test:

first time 46%

second time (one minute after the first time) 62%.


Great! You need at least 25 episodes, but no more than 2 episodes a day, to get the first results. Then the average can be taken into account (with reservations).
 
IlyaA >> :


Great! You need at least 25 episodes to get the first results, but not more than 2 episodes a day. There and the average can be taken into account (with reservations).

You need to develop intuition so that there are results, not chance.

It is better to apply the experience of athletes to this interesting topic.


1. training on the development of abilities needed to provide intuitive thinking (if the intuitive process of processing information can be called thinking).

The goal of this exercise is to have a full mastery of methods of awakening/activation of intuition.

3. a natural and wholesome nutrition. It is necessary to eat as to breathe ;)

After all, the brain and other organs should work perfectly, like Swiss clockwork :)

4. A healthy lifestyle. Well, don't drink, don't smoke - platitudes known to everyone. Besides living in an environmentally friendly place, better outside the city in the countryside, in your own house, surrounded by a lively garden.

5. Positive thinking. Negative thoughts impair the performance of the body.


You should approach the achievement of results energetically and comprehensively.

 
Pavel, bravo!
 

Thank you, Ilya.

It's just my opinion. I believe that this is a kind of pattern (life?) necessary to achieve high results in any business, activity: business, science, politics, sports, art, education and others.


Is there any serious information on the subject (meaning research materials of scientists/candidates for ranks) in Russian?


P.S.: the discussion has gone quiet...

 
meta-trader2007 >> :

Is there any serious information on the subject (meaning research materials of scientists/candidates for ranks) in Russian?


P.S.: something about the discussion has gone quiet...


So the topic is specific. :) And there's work to be done. And there's also a conversation about density in the next thread. You shouldn't expect too many posts here. I have a lot of documentaries, I will share them all. As for doctoral and other scientific research, then there is only communication with acquaintances who have a scientific degree.