Analysis of climate-change-related policies frequently relies on modeling the behavior of energy systems over a long planning horizon under alternative scenarios. There are many taxonomies for classifying these models, with one commonly-used dimension distinguishing between top down models that address energy systems within the context of the larger economy (e.g., computable general equilibrium, or CGE models), versus bottom up models with a higher level of detail on the energy system, but with more limited interactions with the rest of the economy (e.g., partial equilibrium, optimization models such as TIMES/MARKAL). A well-known weakness of both types is their unrealistic depiction of consumer market behavior, which can undermine their credibility with policy makers. A frequent practice when using bottom up models is to address this by imposing ad hoc constraints on consumer behavior (e.g., choice of vehicle, lighting, heating, and cooling technologies) based on “expert judgment,” which, in turn sometimes rely on results from more behaviorally rich discrete choice (or, discrete-continuous choice) models from the literature (DCMs). However, these models assume the same underlying microeconomic-based consumer decision framework on which both top down and bottom up models are also based. By going back to the shared, underlying theory, we have derived theoretical relationships leading to practical methods for directly incorporating consumer preferences from DCMs into pre-existing bottom up modeling systems. We demonstrate this using an empirical application on consumer vehicle choice behavior, where preferences from a nested multinomial logit model in an existing model (MA3T) are incorporated into TIMES/MARKAL.