Article ID: | iaor2001291 |
Country: | South Africa |
Volume: | 24 |
Issue: | 1 |
Start Page Number: | 69 |
End Page Number: | 85 |
Publication Date: | Jan 2000 |
Journal: | International Studies In Economics and Econometrics |
Authors: | Opong K.K., Biepke N., McIlkenny P. |
Keywords: | economics, time series & forecasting methods |
This study uses statistical analysis based on chaos theory to examine the trading volume behaviour of the London Financial Times Stock Exchange (FTSE) All Share Index and FTSE 100 Index. The behaviours of the trading volume series are not independent and identically distributed. The results of the study show that some cycles or patterns occur more frequently than would be expected in a true random series. The results may also explain the observed weak association between stock price changes and trading volume reported in previous studies since those studies were based on linear modelling whereas the behaviour of both stock prices and trading volume may be either a chaotic process, non-linear stochastic process or linear stochastic dependence. A GARCH(1,1) model which appears to explain equity index return series does not explain the behaviour of the equity volume index series on the London Stock Exchange.