Capturing the Impact of Latent Industry-Wide Shocks with Dynamic Panel Model

Expanding the panel model of Pesaran (2006) and Bai (2009), we propose a dynamic panel specification with Bayesian approach to capture the impact of unobservable industry-wide shocks to stock price movements. We employ fundamental accounting information to control company specific shocks and equity market index to capture market wide common shocks. Our model is designed to resolve the potential multicollinearity problem that is known to exist when the industry factors are considered by extracting the industry-wide shocks using Bayesian method.