Factor models commonly used in asset pricing are based on tradable factors that do not represent theoretically relevant risks. To address this issue we develop a factor model that is tightly linked to intertemporal asset pricing theory. Specifically we show that a long-term Bayesian investor prices shocks to the market dividend yield and realized variance as they reflect news to long-term expected returns and volatility. Accordingly we construct intertemporal risk factors as long-short portfolios based on stock exposures to dividend yield and realized variance. We then estimate their risk prices and find that they are consistent with the Intertemporal CAPM under moderate risk aversion. We also show that our intertemporal factor model performs well relative to previous factor models in terms of its tangency Sharpe ratio and its pricing of key test assets including single stocks and industry portfolios.