Visualisation of financial time series by linear principal component analysis and nonlinear principal component analysis

TL;DR

This paper explores how linear and nonlinear principal component analysis can be used to visualize financial time series, aiding technical analysis by highlighting differences during normal and crisis periods.

Contribution

It compares linear and nonlinear PCA methods for financial data visualization and evaluates preprocessing techniques to improve visual insights.

Findings

Nonlinear PCA provides better visualization of crisis periods.

Preprocessing methods significantly affect visualization quality.

Visualizations help distinguish normal and crisis market states.

Abstract

In this dissertation, the main goal is visualisation of financial time series. We expect that visualisation of financial time series will be a useful auxiliary for technical analysis. Firstly, we review the technical analysis methods and test our trading rules, which are built by the essential concepts of technical analysis. Next, we compare the quality of linear principal component analysis and nonlinear principal component analysis in financial market visualisation. We compare different methods of data preprocessing for visualisation purposes. Using visualisation, we demonstrate the difference between normal and crisis time period. Thus, the visualisation of financial market can be a tool to support technical analysis.