Three Essays on Computational Approaches to Economics

Abstract

The three essays included in this dissertation are on three di?erent popular computational approaches that are widely applicable in economics. In Chapter 1, a state-space model is constructed which is linear in state variables and nonlinear in parameters. From the model, the time-varying level of natural interest rate is estimated using Kalman ?lter and Gibbs sampling algorithms. Chapter 2 proposes a new algorithm, called Implicit Particle Gibbs, to solve nonlinear state-space models. And Chapter 3 reviews recent development of deep learning and reinforcement learning algorithms and their applications in economics.