The discovery of what the price of an asset should be constitutes one of the main reasons for the existence of nancial markets. Nowadays this price discovery pro- cess spreads across exchanges because of the prevalence of assets trading in multiple venues. We explore intraday variation in the informativeness of trading venues using a novel framework based on state space methods. Our general approach provides a structure to analyze cases with multiple securities sharing one underlying asset like stocks trading in different exchanges and derivative markets. A simulation study shows our approach is capable of identifying time-variation in information contri- butions. We illustrate the method by applying it to a stock (Expedia) trading in multiple exchanges. The results demonstrate intraday variation in both the relative contribution to price discovery of various exchanges and the total informativeness of the security prices.