Modelling Bid-Ask Spreads in Competitive Dealership Markets

In competitive dealership markets, several transactions may take place within the same time interval so that observed prices cannot be ordered sequentially. Therefore, the standard approach of estimating spread using differenced prices cannot be used without distorting information. We propose an alternative method to model the price behaviour in financial markets for which data differencing is not needed. The model is put into state space form in order to use the Kalman filter smoother to estimate the fundamental price and the spread. The new methodology is illustrated by the transaction data of three liquid stocks on the London Stock Exchange.

Download is not available