Overview
James D. Hamilton is an economist at UC San Diego. He is best known for introducing the Markov-switching (regime-switching) model to econometrics in a landmark 1989 paper on US business cycles, and for the widely used graduate textbook Time Series Analysis (Princeton University Press, 1994). His work spans nonlinear time-series methods, energy economics (oil prices and recessions), and exchange rate dynamics.
Key Contributions / Features
- Markov-switching model (Hamilton 1989) — "A New Approach to the Economic Analysis of Nonstationary Time Series and the Business Cycle," Econometrica 57: 357–384. Modeled US GNP growth as a two-state Markov-switching AR process; developed the Hamilton filter (iterative algorithm for filtered and smoothed regime probabilities); demonstrated that NBER-dated recessions correspond cleanly to the low-growth state. Foundation for the regime-switching literature in econometrics.
- Hamilton filter — Iterative forward pass to compute P(st=j∣y1,…,yt;θ) (filtered probabilities); combined with the Kim (1994) smoother for P(st=j∣y1,…,yT;θ). Core computational tool for all Markov-switching models.
- Long swings in the dollar (Engel-Hamilton 1990) — Applied the MS framework to quarterly dollar exchange rates (DM, franc, pound, 1973–1983); rejected the random walk in favor of stochastic segmented trends for all three currencies; showed that interest rate differentials fail to predict regime transitions, inconsistent with uncovered interest parity. See Uncovered Interest Parity.
- Time Series Analysis (1994) — Standard PhD-level time-series textbook covering unit roots, ARIMA, GARCH, VAR, cointegration, and Markov-switching; one of the most-cited econometrics textbooks.
- Oil prices and recessions — Documented that oil price shocks Granger-cause US recessions; work on energy price uncertainty and Hamilton's nonlinear oil-recession model.
Related