Nathan Palmer


Email address: npalmer4 [at] gmu [dot] edu
Personal website

Nathan Palmer is a Ph.D. candidate in Computational Social Science. He received his B.S in Computer Science from Trinity University in San Antonio and an M.A. in Economics from Boston University. Between educational stints, he spent three years as a Research Assistant at the Federal Reserve Board of Governors, forecasting macroeconomics of the Nordic countries and becoming an expert in FAME.

Nathan strongly believes that agent-based modeling is simply the next step in computational economic modeling, employing insights from software engineering to construct models that "trace out" the effects of "well tested theory" (to use Kydland and Prescott's terminology). The "black box" of agent-based modeling need not exist.

Towards that end, a primary effort of his research is incorporating reinforcement learning, multi-agent learning, and stochastic approximation methods into traditional consumption and asset-pricing frameworks. A goal of this effort is to create agents which are *capable* of attaining an optimal policy in a dynamic stochastic environment, but which are also "tunable" in how quickly they attain an optimal policy.

There are two ways to think about using learning for economic agents: on the one hand, we can ask "what can we attain with the best possible online learning methods?"
On the other hand, we might ask, "what parameterization of this framework matches human learning, and what does that mean for the models in question?"

Aside from attempting to address "traditional puzzles" in macroeconomics, a specific goal of these efforts is to create "portable" boundedly rational agents, who seek to continually do better but may be forever barred from doing so through properties of their environment. Coupled with detailed micro-level data, agents which learn and act dynamically in periods of high uncertainty may be very useful in large-scale, detailed models of macroeconomic and financial systems, especially for answering questions about short-term, out-of-equilibrium dynamics.

In his spare time, Nathan tries to cook Tex-Mex and enjoy good friends' company. He wishes that he hiked and danced as much as he used to, but life is busy these days...