Discussing the article: "Singular Spectrum Analysis in MQL5"

[MetaQuotes]

Check out the new article: Singular Spectrum Analysis in MQL5.

This article is meant as a guide for those unfamiliar with the concept of Singular Spectrum Analysis and who wish to gain enough understanding to be able to apply the built-in tools available in MQL5.

Recent iterations of MetaTrader 5 have introduced the initial integration of OpenBLAS methods into its core vector and matrix data types. Of particular interest are a set of methods related to Singular Spectrum Analysis (SSA). In this article, we explore the new tools available in MQL5 related to SSA and unpack how they can be used in analysis and forecasting. This guide aims to provide a resource for traders seeking to harness SSA's full potential. We will delve into the core SSA methodology, demystifying its two-stage decomposition and reconstruction process. More importantly, we will discuss what each of the new SSA vector methods do and demonstrate how to interpret and optimally combine their outputs for actionable insights.

Singular Spectrum Analysis is a non-parametric technique designed for the analysis and forecasting of time series data. Its objective is to decompose a time series into a few additive components, which typically include a slowly varying trend, various cycles, and residual noise. A distinguishing feature of SSA is its minimal reliance on predetermined assumptions regarding the underlying data-generating process. The conceptual foundation of SSA integrates elements from statistics and signal processing. Essentially, SSA is rooted in spectral decomposition, allowing it to reveal the frequency characteristics of a time series by analyzing its principal components in a reconstructed embedding space. It can be effectively conceptualized as a form of Principal Component Analysis (PCA) specifically adapted for time series data, leveraging the principles of dimensionality reduction to uncover hidden structures and patterns that might be obscured by noise or complex interactions.

Author: Francis Dube

[Stanislav Korotky]

It's strange to see so closely related articles on the same topic (even if one of them was originally written in Russian) in very short period of time.

Articles

One-dimensional singular spectrum analysis

Evgeny Chernish , 04/23/2025 11:23

The article examines the theoretical and practical aspects of the singular spectrum analysis (SSA) method, which is an effective method for analyzing time series that allows the complex structure of a series to be represented as a decomposition into simple components such as trend, seasonal (periodic) fluctuations, and noise.