search query: @indexterm FORECASTING / total: 1417
reference: 157 / 1417
| Author: | Tolvi, J. |
| Title: | Outliers and predictability in monthly stock market index returns |
| Journal: | Liiketaloudellinen aikakauskirja
2002 : 4, p. 369-380 |
| Index terms: | Stock markets Forecasting Models |
| Language: | eng |
| Abstract: | The predictability of stock market returns, of either individual stocks or indices, is an old research topic. Three simple statistical models are used in this study for predicting stock market index returns. A random walk model is used as benchmark, which any other model should be able to improve on. A basic autoregressive (AR) model is then compared with an AR-outlier model,where dummy variables are added for detected outliers. It is assumed that if such out liers are taken into account in the model, the predictions should become better compared to the basic AR model. The data used consists of monthly stock market indices from 15 OECD countries. The results indicate among others that for the autocorrelated series, taking outliers into account in making predictions will indeed improve the one step ahead predictions. |
SCIMA