We consider two players facing identical discrete-time bandit problems with a safe and a risky arm. In any period, the risky arm yields either a success or a failure, and the first success reveals the risky arm to dominate the safe one. When payoffs are public information, the ensuing free-rider...

We consider two players facing identical discrete-time bandit problems with a safe and a risky arm. In any period, the risky arm yields either a success or a failure, and the first success reveals the risky arm to dominate the safe one. When payoffs are public information, the ensuing free-rider...

This paper studies a game of strategic experimentation with two-armed bandits whose risky arm might yield a payoff only after some exponentially distributed random time. Because of free-riding, there is an inefficiently low level of experimentation in any equilibrium where the players use...

There is strong evidence that macroeconomic releases influence prices in financial markets. However, why do markets react to some announcements while they ignore others with a similar content? Based on a Bayesian learning model, we show that market impact is mainly determined by information...

An important claim of Bayesian learning and a standard assumption in price discovery models is that the strength of the price impact of unanticipated information depends on the precision of the news. In this paper, we test for this assumption by analyzing intra-day price responses of CBOT T-bond...

We analyze how markets adjust to new information when the reliability of news is uncertain and has to be estimated itself. We propose a Bayesian learning model where market participants receive fundamental information along with noisy estimates of news' precision. It is shown that the efficiency...

This paper studies a game of strategic experimentation in which the players have access to two-armed bandits where the risky arm distributes lumpsum payoffs according to a Poisson process with unknown intensity. Because of free-riding, there is an inefficiently low level of experimentation in...

We establish convergence of beliefs and actions in a class of one-dimensional learning settings in which the agent's model is misspecified, she chooses actions endogenously, and the actions affect how she misinterprets information. Our stochastic-approximation-based methods rely on two crucial...