We develop a framework for estimating expected returns—a predictive system—that allows predictors to be imperfectly correlated with the conditional expected return. When predictors are imperfect, the estimated expected return depends on past returns in a manner that hinges on the correlation between unexpected returns and innovations in expected returns. We find empirically that prior beliefs about this correlation, which is most likely negative, substantially affect estimates of expected returns as well as various inferences about predictability, including assessments of a predictor's usefulness. Compared to standard predictive regressions, predictive systems deliver different expected returns with higher estimated precision.

[We develop a model of the optimal initial public offering (IPO) decision in the presence of learning about the average profitability of a private firm. The entrepreneur trades off diversification benefits of going public against benefits of private control. Going public is optimal when the firm's expected future profitability is sufficiently high. The model predicts that firm profitability should decline after the IPO, on average, and that this decline should be larger for firms with more volatile profitability and firms with less uncertain average profitability. These predictions are supported empirically in a sample of 7183 IPOs in the United States between 1975 and 2004.]

We develop a general equilibrium model in which stock prices of innovative firms exhibit “bubbles” during technological revolutions. In the model, the average productivity of a new technology is uncertain and subject to learning. During technological revolutions, the nature of this uncertainty changes from idiosyncratic to systematic. The resulting bubbles in stock prices are observable ex post but unpredictable ex ante, and they are most pronounced for technologies characterized by high uncertainty and fast adoption. We find empirical support for the model's predictions in 1830–1861 and 1992–2005 when the railroad and Internet technologies spread in the United States.

We develop a framework for estimating expected returns—a predictive system —that allows predictors to be imperfectly correlated with the conditional expected return. When predictors are imperfect, the estimated expected return depends on past returns in a manner that hinges on the correlation between unexpected returns and innovations in expected returns. We find empirically that prior beliefs about this correlation, which is most likely negative, substantially affect estimates of expected returns as well as various inferences about predictability, including assessments of a predictor's usefulness. Compared to standard predictive regressions, predictive systems deliver different expected returns with higher estimated precision.

We develop a model of the optimal initial public offering (IPO) decision in the presence of learning about the average profitability of a private firm. The entrepreneur trades off diversification benefits of going public against benefits of private control. Going public is optimal when the firm's expected future profitability is sufficiently high. The model predicts that firm profitability should decline after the IPO, on average, and that this decline should be larger for firms with more volatile profitability and firms with less uncertain average profitability. These predictions are supported empirically in a sample of 7183 IPOs in the United States between 1975 and 2004. The Author 2008. Published by Oxford University Press on behalf of The Society for Financial Studies. All rights reserved. For Permissions, please email: [email protected], Oxford University Press.