Historically economists have relied on econometric (or statistical) methods to estimate parameters from observed data. In this approach we observe a rich cross-section or time-series dataset, specify an economic model which implicitly defines the underlying behavior (say, simple linear regression), and estimate key parameters of interest (such as supply and demand elasticities). Econometric analysis typically requires a large dataset and we will often specify a reduced-form model. What do we do when data are limited? What do we do when we want to predict response to policies that simultaneously affect multiple resources, production activities, prices, and markets? Computational methods such as linear programming, calibrated optimization, and dynamic programming allow us to calibrate parameters using limited data and specify a framework that is consistent with economic theory that we can then use to simulate the interaction of complex resource policies.