Bivariate Nonparametric Density Estimation of Stock Prices and Volume

This paper comprehensively investigates the joint movement of stock prices and trading volume of New York and Tokyo stock markets by undertaking nonparametric density estimation. Bivariate nonparametric density estimation has been reported as a powerful tool for revealing complicated relations between two variables. In application to finance, it is important to use a method robust for heavy-tailed densities, since the distributions of asset price changes are known to have heavy tails, and information about sudden and large price changes is contained in the tails. The empirical regularities found in this paper are mostly consistent with previous literature, but partially disagrees with the work of Gallant et al. (1992).