Modeling and forecasting the intraday volume of Shanghai security composite index

In this paper, we put forward a modeling method for the intraday volume series that exhibits W shape structure and heteroskedasticity. Firstly, we take the averaged values of historical intraday volume as the periodic component of volume. Then we choose proper mean-variance models to simulate the volume series removed intra-daily periodic trend. Finally, we use the period-mean-variance model selected to predict future volume. By empirical analyzing the Shanghai security composite index on 59 days with time interval of 1 minute, we demonstrate that with this method we can effectively extract periodic components in volume series and ARMA-EGARCH (2, 2) outperform others in stimulation and forecast for the intraday volume time series.