Check out some of my projects
Using data from 387 households in rural Kenya, we model source choice and water demand using a discrete-continuous (linked) demand model.
A decision support tool for rural water supply that planners or community members can use to simulate scenarios such as (1) price, quality, or placement changes of existing sources, (2) the closure of an existing source, or (3) the addition of a new source. The tool is freely-available on the web, and draws from empirical research on household water source choice and demand studies.
This project uses data from the American Housing Survey, and exploits variation in payment-status (who pays the energy bill), to estimate the effect of information asymmetries on the adoption of efficient (Energy Star rated) technologies in the U.S. rental housing market.
In this paper I develop a reduced form multiple discrete-continuous demand model to characterize demand for scenarios in which consumers face two distinct, but related, decisions: which goods to consume, and (of the goods that are consumed) in what quantities (e.g. which stocks to own and how many shares of each).
Predicted rental housing prices using machine learning methods to estimate the rent premium of utility-included rental contracts. (Includes applications of hyperparameter tuning, and optimization over neural network architectures using random search.)