Discussing the article: "MQL5 Wizard Techniques you should know (Part 12): Newton Polynomial"

[MetaQuotes]

Check out the new article: MQL5 Wizard Techniques you should know (Part 12): Newton Polynomial.

Newton’s polynomial, which creates quadratic equations from a set of a few points, is an archaic but interesting approach at looking at a time series. In this article we try to explore what aspects could be of use to traders from this approach as well as address its limitations.

Time series analysis plays an important role not just in supporting fundamental analysis but in very liquid markets like forex, it can be the main driver for decisions on how one is positioned in the markets. Traditional technical indicators have tended to lag the market a lot which has brought them out of favor for most traders, leading to the rise of alternatives perhaps the most predominant of which, at the moment is neural networks. But what about polynomial interpolation?

Well they present some advantages mainly from being easy to understand and implement since they explicitly present the relationship between past observations and future forecasts in a simple equation. This helps in understanding how past data impacts future values which in turn leads to developing broad concepts and possible theories on the studied time series’ behavior.

In addition, being adaptable to both linear and quadratic relations make them flexible to various time series and perhaps more pertinent for traders, capable of coping in different market types (e.g. ranging vs trending or volatile vs calm markets)

Furthermore, they are typically not compute-intense and are relatively lightweight when compared to alternative approaches like neural networks. In fact, the model(s) examined in this article have zero storage requirements the kind you would need with say a neural network where depending on its architecture where making provision for storing a lot the optimal weights and biases after each training session is a requirement.

Author: Stephen Njuki

[linfo2]

Thank you Stephen , Very interesting subject and well written .Is there supposed to be  Cnewton.mqh in the downloads?. I get Cnewton.mqh' not found SignalWZ_12.mqh ,it seems to be referred to in all 3 examples

[linfo2]

linfo2 #:
Thank you Stephen , Very interesting subject and well written .Is there supposed to be  Cnewton.mqh in the downloads?. I get Cnewton.mqh' not found SignalWZ_12.mqh ,it seems to be referred to in all 3 examples

Thank you for your Ideas Stephen , I am now looking for other ways to use this the Newton Polynomial much appreciated.