Price-volume relation in stocks: A multiple time series analysis on the Singapore market

We examine the price-volume relation in stocks using the multiple time series approach due to Tiao and Box (1981). This approach has the advantage of treating price and volume jointly and symmetrically (without enforcing the roles of input and output). It is free of the simultaneity bias in regression analysis and the unidirectional dynamics imposed by transfer function models. Empirical results show that there is implicit positive correlation between price and volume through their residuals. However, the results for the explicit lead and lag relations are mixed. The technical analysts' adage that volume often leads the trend of price is not supported. Nonetheless, the implicit relationship between price and volume confirms the usefulness of incorporating volume data to forecast future return. Our analysis shows that the multiple time series models outperform the univariate models in post-sample forecasts.