Abstract
This study investigates the characteristic of non-stationarity in a financial time-series and its effect on the learning process for Artificial Neural Networks (ANN). It is motivated by previous work where it was shown that non-stationarity is not static within a financial time series but quite variable in nature. Initially unit-root tests were performed to isolate segments that were stationary or non-stationary at a pre-determined significance level and then various tests were conducted based on forecasting accuracy. The hypothesis of this research is that when using the de-trended/original observations from the time series the trend/level stationary segments should produce lower error measures and when the series are differenced the difference stationary (non-stationary) segments should have lower error. The results to date reveal that the effects of variable stationarity on learning with ANNs are a function of forecasting time-horizon, strength of the linear-time trend, sample size and persistence of the stationary process.
| Original language | English |
|---|---|
| Title of host publication | 2011 IEEE Symposium on Computational Intelligence for Financial Engineering and Economics (CIFEr) |
| Place of Publication | Paris |
| Publisher | IEEE |
| Pages | 1 -8 |
| Number of pages | 8 |
| ISBN (Print) | 978-1-4244-9933-5 |
| DOIs | |
| Publication status | Published - 1 Apr 2011 |
Keywords
- artificial neural network
- detrended original observation
- difference stationary segments
- financial time series
- learning process
- time horizon forecasting
- trend level stationary segments
- unit root tests
- finance
- learning (artificial intelligence)
- neural nets
- time series