Discussion of article "Forecasting Time Series (Part 2): Least-Square Support-Vector Machine (LS-SVM)"
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
New article Forecasting Time Series (Part 2): Least-Square Support-Vector Machine (LS-SVM) has been published:
This article deals with the theory and practical application of the algorithm for forecasting time series, based on support-vector method. It also proposes its implementation in MQL and provides test indicators and Expert Advisors. This technology has not been implemented in MQL yet. But first, we have to get to know math for it.
Let us run a single test.
EA LSSVMbot Report on XAUUSD D1, 2017-2020
Not really amazing performance, but basically, the system works. Date ranges are marked on the report chart, from which the training data was taken to find optimal "gamma" and "sigma" (highlighted in green),which range was defined in the tester in training mode (highlighted in yellow), and the range where the EA traded on unknown data (highlighted in pink).
The ways of interpreting the forecast and constructing a trading strategy around it can be different. In particular, in our test EA, there is an input, PreviousTargetCheck (false, by default). It being enabled, the forecast-based trading will be performed using another strategy: Transaction direction is determined by the location of the newest forecast relative to the preceding one. There is also some further scope for experimenting with other settings, such as SOM clusterization, changing the lot size depending on the strength of the forecasted movement, refilling, etc.
Author: Stanislav Korotky