Discussion of article "Universal Regression Model for Market Price Prediction" - page 7

New comment
 
Integer:

There should be formulas in the excel help.

Yeah. There are.

 
As far as I understand, the proposed regression model is another way of extrapolating the course by smooth functions. Question: What are the advantages of the regression model over simple extrapolation by polynomials or, for example, SSA caterpillar?
 

Virty:
Насколько я понял, предложенная регрессионная модель есть ещё один способ экстраполяции курса гладкими функциями. Вопрос: Каковы преимущества регрессионной модели по сравнению с простой экстраполяцией полиномами или, например, гусеницей SSA?

1.  Как показывают исследования, приведенные в статье, Вы правы отчасти, поскольку предложенная регрессионная модель, действительно есть ещё идин, новый способ экстраполяции курса не всякими гладкими функциями, а именно разновидностями Гамма-функции Эйлера, известными науке как плотность(двухпараметрической) функции распределения Эрланга- функция  М(t/т;n+1;1;1) , интегральная (двухпараметрическая) функция распределения Эрланга-функция  S(t/т;n+1;1;1) и новой, неизвестной науке,  выявленной и названной мной, как интегральная (двухпараметрическая) экспоненциальная функция распределения - L(t/т;n;1;1), превращающаяся, при подстановки в нее значения параметра n=1, в известную интегральную (однопараметрическую) экспонециальную функцию распределения.

2. Obvious (we will come back to this issue later)

 
Virty:
As far as I understand, the proposed regression model is another way to extrapolate the course by smooth functions. Q: What are the advantages of the regression model over simple extrapolation by polynomials or, for example, SSA caterpillar?

Exclusively solid disadvantages over polynomials - you can get the coveted polynomial order in them.... It is always significant - the correlation of values, not with what and how the series are approximated (here you can even visually see that the correlation of the forecast to the fact is at zero, so all this is not suitable for use...) and it is customary to choose the simplest method....
 
yosuf:

1.According to my newly proposed vision.

This "new vision" with a pompous title is nothing but a terry ignorance, lack of knowledge of the relevant literature for the last 100 or 200 years. If the author had read the simplest textbooks on matstatistics or econometrics, he would not have published an article on forecasting without using the words "hypothesis test" and"confidence interval" in this article. A prediction is made without stating confidence in the model itself, and without stating confidence in the prediction. Another nerd blizzard.
 
dasmen:
Exclusively solid disadvantages before polynomials - in them you can get the desired order of the polynomial.... What is always significant is the correlation of values, not what and how the series are approximated (here you can even visually see that the correlation of the forecast to the fact is at zero, so all this is not usable...) and at the same time it is customary to choose the simplest method....
Please choose any piece of the market, exactly the market, not a sequence of invented figures, process it with a polynomial with the desired order, achieve correlation of values so that even visually it was not visible that the correlation of thethe correlation between the forecast and the fact is at zero, pass the initial data to me and let's compare the obtained forecast results with their actual results, which really turned out in the market conditions following the initial data, but everyone is ready to assert verbatim and such a method of critical evaluation is not welcome in such complicated situations as market analysis.
 
faa1947:
This "new vision" with a pompous title is nothing but blatant ignorance, lack of knowledge of the relevant literature for the last 100 or 200 years. If the author had read the simplest textbooks on matstatistics or econometrics, he would not have published an article on forecasting without using the words "hypothesis test" and "confidence interval" in this article. A prediction is made without stating confidence in the model itself, and without stating confidence in the prediction. Another nerd blizzard.
Within the limits of this article I did not set the tasks described by you, so that it would not turn into a multi-page monograph, within the limits of which I intend to cover not only the issues mentioned by you. Based on the conclusions of the article, I have already developed 8 brand new indicators - Forecast 001-006, SultonovPrediction 1 and 2, three Expert Advisors - eForell 01-03, eForell 01-03, and eForell 01-03. eForell 01-03, taking into account both qualitative and quantitative characteristics of the market according to the ratio (18) of the article, which are actively discussed on the forum mql4 Sultonov Indicator on the MT screen, create an Expert Advisor for mt4 on the programme made on exel, Creating a trading robot.
Индикатор Султонова - MQL4 форум
  • www.mql5.com
Индикатор Султонова - MQL4 форум
 
yosuf:
Within the limits of this article I did not set the tasks described by you, so that it did not turn into a multi-page monograph, within the limits of which I intend to cover not only the issues mentioned by you. Based on the conclusions of the article, I have already developed 8 brand new indicators - Forecast 001-006, SultonovPrediction 1 and 2, three Expert Advisors - eForell 01-03, eForell 01-03, and eForell 01-03. eForell 01-03, taking into account both qualitative and quantitative characteristics of the market according to the ratio (18) of the article, which are actively discussed on the forum mql4 Indicator Sultonov on the MT screen, create an Expert Advisor for mt4 on the programme made on exel, Creating a trading robot.
I looked through the forum. in the very beginning you were referred to GOST, but you continue to bend your own, as they say, to each his own.
 
yosuf:
Within the limits of this article I did not set the tasks described by you, so that it did not turn into a multi-page monograph, within the limits of which I intend to cover not only the issues mentioned by you. Based on the conclusions of the article, I have already developed 8 brand new indicators - Forecast 001-006, SultonovPrediction 1 and 2, three Expert Advisors - eForell 01-03, eForell 01-03, and eForell 01-03. eForell 01-03, taking into account both qualitative and quantitative characteristics of the market according to the ratio (18) of the article, which are actively discussed on the forum mql4 Indicator Sultonov on the MT screen, create an Expert Advisor for mt4 on the programme made on exel, Creating a trading robot.
Good day dear Yusufkhodja, can you tell me where I can download the indicators and Expert Advisors you are talking about?
 
lyvel:
Good day, dear Yusufhoja, where can I download the indicators and Expert Advisors you are talking about?
Contact me via private message, you will find out about the conditions.
12345678
New comment