Published as Opuscula ISRN HEV-BIB-OP-30-SE
|
Abstract:
This paper presents statistical investigations regarding the predictability of stock returns. The examined data covers 207 stocks on the Swedish stock market for the time period 1987-1996. The results show trend behavior and autocorrelation values that are stable even when the entire time interval is broken down to yearly intervals. A proposed concept of daily returns RD and a comparision to the ordinary step returns R S, show a significant difference in the data, caused by the holiday effect. It is also shown that seasonal variables, such as the month of the year, affect the stock returns more than the average daily changes. This is consequential for all methods where the seasonal variables are not taken into account when predicting daily stock returns.
Citations
| 55 | Predicting sunspots and exchange rates with connectionist networks, In: Nonlinear Modeling and Forecasting – Weigend, Huberman, et al. - 1992 |
| 51 | Efficient Capital Markets: II – Fama - 1991 |
| 51 | Economic prediction using neural networks: the case of the IBM daily stock returns – White - 1998 |
| 13 | On the predictability of common stock returns: Worldwide evidence – Hawawini, Keim - 1995 |
| 10 | Time series prediction by adaptive networks: A dynamical systems perspective – Lowe, Webb - 1991 |
| 6 | Market inefficiencies, technical trading and neural networks – Baestaens, Bergh, et al. - 1996 |
| 5 | Testing the efficent markets hypothesis with gradient descent algorithms – Tsibouris, Zeidenberg - 1995 |
