Discussing the article: "Cycles and trading"

New comment
 

Check out the new article: Cycles and trading.

This article is about using cycles in trading. We will consider building a trading strategy based on cyclical models.

The main task facing a trader is to predict price movements. Traders build their forecasts based on one model or another. One of the simplest and most visual models is the cyclical price movement model.

The basic idea behind any cyclical pattern is that various factors interact to create cycles in price movement. These cycles may differ from each other in their duration and strength. If you know the parameters of these cycles, then trading operations will become very simple: open a buy position when the cycle has reached its minimum, sell when the cycle has reached its maximum.

Let's see how this model can be used in practice.


Author: Aleksej Poljakov

 
Please clarify, please, for especially clever guys:

Is it after optimisation results? Or in a "head-on" straightaway? (forward)
 
Ivan Butko #:
Please clarify, please, for especially clever guys:

Is it after optimisation results? Or in a "head-on" straightaway? (results)

the first 2 are optimisation, and all the others - I poked around and then tried the parameters +/- left what I liked best.

 
Aleksej Poljakov #:

The first 2 - optimisation, and all the others - I poked around then tried the parameters +/- left what I liked more

Thank you

 
Ivan Butko #:

Thank you

There are other possibilities - instead of prices use their logarithms. The time series becomes smoothed and you can throw away all the SMAs and so on.

The second is to use differences of different degrees of accuracy... but, here I'm racking my brain - how to explain it simply.

 
At the first quick reading the material seemed interesting.... I'll reread it and see what's on the auction... give you feedback here. Thank you.
 
In impulse equilibrium theory, there is a concept of M-cyclicity.
 
Its a great article, but it is also a Cookie Jar, so many alternatives to sample.  Have you considered creating an adaptive engine that evaluates the market conditions and attempts to select the best optimization for the current conditions
 
CapeCoddah #:
This is a great article, but it's also a biscuit jar, so many alternatives to try. Have you considered building an adaptive mechanism that evaluates market conditions and tries to choose the best optimisation for current conditions

One way to adapt is to evaluate multiple cycles at once. Moreover, this is easier than it seems. For example, you can do it this way. The first cycle - take counts in a row. The second cycle - take the price samples one after another. And so on. The combination of these cycles will give a unique picture of the market state at the moment.

 
Thanks for the great suggestion.  I will try it and let you know but it will be a while.
 

🚫 Red Flags:

  1. Unusable default parameters:

    • iPeriod = 870 , R = -940 , S = 450 → absurd values for short-term trading

  2. No trades triggered:

    • The EA evaluates the signal only once per new bar, and signal logic thresholds almost never hit with default parameters.

  3. CalcLWMA() uses static accumulators in the original — causing totally invalid results over time.

  4. No backtest or validation in code — and the indicator isn’t provided in the article for real-time visual inspection.

  5. Boasts of equity growth without sharable evidence or MQ5 Signals links.

12
New comment