Predicting Intraday Financial Market Dynamics Using Takens' Vectors; Incorporating Causality Testing and Machine Learning Techniques
Traditional approaches to predicting financial market dynamics tend to be linear and stationary, whereas financial time series data is increasingly nonlinear and non-stationary. Lately, advances in dynamical systems theory have enabled the extraction of complex dynamics from time series data. These...
| Main Author: | |
|---|---|
| Format: | Others |
| Language: | English |
| Published: |
Digital Commons @ East Tennessee State University
2015
|
| Subjects: | |
| Online Access: | https://dc.etsu.edu/etd/2582 https://dc.etsu.edu/cgi/viewcontent.cgi?article=3965&context=etd |
| Summary: | Traditional approaches to predicting financial market dynamics tend to be linear and stationary, whereas financial time series data is increasingly nonlinear and non-stationary. Lately, advances in dynamical systems theory have enabled the extraction of complex dynamics from time series data. These developments include theory of time delay embedding and phase space reconstruction of dynamical systems from a scalar time series. In this thesis, a time delay embedding approach for predicting intraday stock or stock index movement is developed. The approach combines methods of nonlinear time series analysis with those of causality testing, theory of dynamical systems and machine learning (artificial neural networks). The approach is then applied to the Standard and Poors Index, and the results from our method are compared to traditional methods applied to the same data set. |
|---|