This paper develops a methodology to estimate models of demand for differentiated durable goods with fully heterogeneous consumers that extends standard estimation techniques to account for the dynamic concerns of consumers. A “nested” technique is used to uncover a reduced form of the solution to the dynamic optimization problem of consumers which can be easily incorporated into a conventional discrete choice framework. The presence of such reduced form serves as the control for the dynamic optimization problem and allows a consistent estimation of the deep preference parameters, which if desired can be used to compute the complete dynamic problem. A simple application to the market for digital cameras is used to demonstrate the feasibility of the technique. The methodological framework has potential applicability in other dynamic demand problems. It is also substantially less costly than alternative estimation approaches, which in principle require the computation of the dynamic optimization problem across all consumer types and through all the steps of the estimation algorithm.