FEPA — An integrated computational intelligence model for predicting financial time series
| Autor: | Heping Pan, Yu Ma, Chengzhao Zhang |
|---|---|
| Rok vydání: | 2017 |
| Předmět: |
Finance
Artificial neural network business.industry Computer science Dimensionality reduction 05 social sciences Feature extraction 050301 education Computational intelligence Hilbert–Huang transform Sliding window protocol 0502 economics and business Principal component analysis Time series business 0503 education 050203 business & management |
| Zdroj: | SII |
| DOI: | 10.1109/sii.2017.8279187 |
| Popis: | This paper presents an integrated computational intelligence model — FEPA — for predicting financial time series (FTS), integrating Empirical Mode Decomposition (EMD) for signal processing, Principal Component Analysis (PCA) for dimension reduction and feature extraction, and Artificial Neural Networks (ANN) for prediction modeling. The model uses a sliding window to capture the most recent time series data, applies EMD to transform the data into multilevel Intrinsic Mode Functions (IMF's), uses PCA to reduce the dimensionality of IMF's and to generate a set of information-rich features which are input into an ANN to generate the prediction. This work is original in four aspects: 1) a structural reformulation of EMD algorithm, 2) a sliding window for tackling the end effect of EMD, 3) investigation of multi-step prediction, 4) testing on two levels of time frame: D1 (daily) and M15 (15-minutely). The new model is tested on the historical data of two stock indices — Chinese HS300 and Australian AXJO, achieving a hit rate of 78% and 82% on HS300 D1 and M15, and 74% on AXJO D1 respectively. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|