Discussion of article "Universal Regression Model for Market Price Prediction"

 

New article Universal Regression Model for Market Price Prediction is published:

The market price is formed out of a stable balance between demand and supply which, in turn, depend on a variety of economic, political and psychological factors. Differences in nature as well as causes of influence of these factors make it difficult to directly consider all the components. This article sets forth an attempt to predict the market price on the basis of an elaborated regression model.


Author: Юсуфходжа

 

Ready to ask the first question.

The very first formula in the article is justified in this way:

Предположим далее, что в течении бесконечно малого отрезка времени dt воздействующая сила уменьшится на величину dD(t) пропорционально оставшейся к моменту времени t силе D(t):

With this formulation, there must necessarily be a minus (force decreases) in front of the right-hand side of the equation. And there isn't...

Although this is rather a typo - the equation is solved as if the minus were there.

 
alsu:

I'm ready to ask the first question.

The very first formula in the article is justified as follows:

With this formulation, there must necessarily be a minus (force decreases) in front of the right-hand side of the equation. And there isn't...

Although this is rather a typo - the equation is solved as if the minus were there.

It is a typo, we will ask the moderator to correct it.
 
alsu:

Ready to ask the first question.

ready to ask the second question ;)

why didn't you add Excel calculations to the article ? secret intention, or did you forget ?

I would like to check the results, but alas, I don't remember the maths anymore, and I'm afraid to make a mistake when writing the code, I've long since passed higher maths and forgotten it for uselessness, it's easier for me now to knock on my laptop or mount/service controllers.

 
yosuf:

Could you please post an excel that already has all the equations you are describing?
 

Model upon model. Is there anything of the reality that people live in? Real statistics that confirm the properties of the model?

 
IgorM:

ready to ask the second question ;)

why didn't you add Excel calculations to the article ? secret intention, or did you forget ?

I would like to check the results, but alas, I don't remember the maths anymore, and I'm afraid to make a mistake when composing the code, I've long passed higher maths and forgotten it for uselessness, it's easier for me now to tap on my laptop or mount/service controllers.

so as not to overload the article and not to distract from the essence of the discussion of purely theoretical issues. This article is intended to highlight and justify the validity of the theory, but the necessary amount of actual testing is given in the article.
 
yosuf:

Now the regression equation (10b), for predicting the market price P(t), takes the following final form:

Where P(0)corr and Dcorr contain information about the sum of the actual price values and the trend direction of the actual data for t = 0,1,2,......k;

Does your formula assume forecasting for t > k?

I assume you have not in any way compared your regression prediction with the linear regression prediction for t > k?

 
sergeev:
Can you please post an excel that already has all the equations you are describing?
Why am I obliged to post exel? If you mean my suggestion on the forum to find developers to translate excel to mql4 for MT4, then it was meant that further conversation will go on a confidential level, and this suggestion remains in force, especially I made a statement known to you today at 9.00 Moscow time.
 
yosuf:
estimated and forecast (P1) - (red line)
According to the legend of the above graph, P1 is a dark blue smooth curve. Is it somewhere from the 38th count showing the forecast?
 
Vita:

Where P(0)corr and Dcorr contain information about the sum of the actual price values and the trend direction of the actual data for t = 0,1,2,......k;

Does your formula assume forecasting for t > k?

I assume you have not in any way compared your regression prediction with the linear regression prediction for t > k?

k is a sample size that each researcher chooses for himself and I don't understand the t>k question, am I limiting k? take as much as you want.