We develop and estimate a model of transaction price change with asymmetric information and inventory costs. One special feature of our model is that it captures not only the effect of the market-maker's inventory on stock prices but also that of large traders. Moreover it takes into account the interactions between inventory control and information trading. We look at the implications of our model concerning the impact of the order flow and its expected component on the price, and compare them with existing models. Using transactions data for a sample of NYSE stocks, we reject the specifications of early empirical studies of asymmetric information or inventory control models. As for more recent papers -including ours- that account for both effects, our conclusion depends on the regression method we consider. We find that we should either reject their specifications or conclude that this rejection is actually due to our choice of a -commonly used- proxy for the expected order flow.
CHAU, M. and LIM, J. (2000). The Dynamic Response of Stock Prices under Asymmetric Information and Inventory Costs: Theory and Evidence. ESSEC Business School.