This paper investigates the performance of early warning systems for currency crises in real-time, using forecasts of indicators that are available at the moment predictions are to be made. We focus on eight Latin American and Central and Eastern European countries, distinguishing an estimation period 1990{2009 and a prediction period 2010{2014. We apply two varieties of early warning systems: the signal approach and the logit model. For both
methods we nd that using early estimates in the predictions worsens the ability of early warning systems to signal crises compared to the most recently available information.