his article empirically examines the determinants of volume in metal futures T markets. First, a theoretical model is developed. Based on the assumption that the trading agents can be divided into two groups, speculators and hedgers, it is found that volume can be represented as a function of inter-and intraday price volatilities and an information set. The results indicate that in both the daily and the monthly regression equations, the most important variables are the volatility measures with interest rates, open interest, and inflation also recording expected results.
I. INTRODUCTION
This study concerns the relationship between the level of activity in the metal futures markets and macro-and microeconomic variables. Previous studies, such as Cornell (1981), and Tauchen and Pitts (1983), among others, concentrated on the volume-price variability relationship in both equity and futures markets. Very little attention, in the studies above or other related studies, is given to variables other than interday price variability that might affect the level of activity in equity and futures markets based on the fact that there are two main groups of traders, hedgers and speculators.'
This article attempts to fill this gap by systematically examining the determinants *The opinions expressed in this article are not necessarily those of COMEX.
'Grossman (1977) is an exception to this general statement. In a theoretical model he suggests that inventory also might be a variable that significantly influence volume levels. He further suggests that the volume of trade in the market should be positively related to the level of inventory. To our knowledge, his theoretical claim has never been tested empirically.