We propose the dynamic network effect (DNE) model for the study of high-dimensional multivariate time series data. Cross-sectional dependencies between units are captured via one or multiple observed networks and a low-dimensional vector of latent stochastic network effects. The...

We investigate the benefits of forecast combination for timing equity factors based on predictive regressions using macro predictors. Relative to standard predictive regression models, forecast combination reduces the noise of forecasts and hence improves their out-of-sample predictive accuracy....

In many environments, including credit and online markets, past records about participants are collected, published, and erased after some time. We study the effects of erasing past records on trade and welfare in a dynamic market where each seller's quality follows a Markov process and buyers...

This paper attempts to assemble evidence for the relationship between the product and the financial market. Drawing back on work in industrial organization, we analyze the relationship between profit persistence and expected stock returns. We show that long-run profit persistence together with...