Overview
Charles R. Nelson is an economist at the University of Washington (Seattle). He is best known for the Nelson-Plosser (1982) paper establishing that most U.S. macroeconomic time series contain unit roots, and for the graduate textbook State-Space Models with Regime Switching (with Chang-Jin Kim, MIT Press 1999) which is a standard reference for Bayesian Markov-switching time-series analysis.
Key Contributions / Features
- Nelson-Plosser (1982) — "Trends and Random Walks in Macroeconomic Time Series: Some Evidence and Implications," Journal of Monetary Economics 10: 139–162. Applied Dickey-Fuller tests to 14 U.S. macroeconomic series; found 13 of 14 are better described as difference-stationary (random walks) than as trend-stationary processes. One of the most cited papers in macroeconometrics; established that ARIMA modeling is more appropriate than detrending for most macro aggregates.
- Business cycle turning points and coincident index (Kim-Nelson 1998) — With Chang-Jin Kim, combined the Stock-Watson (1989, 1991) dynamic factor model with Hamilton's (1989) Markov-switching model and estimated via multimove Gibbs sampling. Generated a new experimental coincident index from four DOC monthly indicators and tested for business cycle duration dependence via Bayesian variable selection. Strong evidence of positive duration dependence in recessions; weak evidence for booms. See Kim-Nelson (1998).
- Plucking model econometrics (Kim-Nelson 1999a) — With Chang-Jin Kim, provided the first formal econometric model of Friedman's (1964) plucking hypothesis: a state-space UC model with an asymmetric Markov-switching transitory shock, estimated via Kim's (1994) approximate MLE on U.S. real GDP and unemployment. Finds strong evidence that GDP recessions are transitory downward plucks from a stochastic trend ceiling. See Plucking Model and Kim-Nelson (1999a).
- State-Space Models with Regime Switching (Kim and Nelson 1999) — Graduate textbook covering the Kim filter and smoother, Gibbs sampling for Markov-switching state-space models, and empirical applications to business cycles, exchange rates, and interest rates.
- Great Moderation and structural break in MS model (Kim-Nelson 1999b) — With Chang-Jin Kim, identified a structural break at 1984:Q1 in Hamilton's (1989) Markov-switching business-cycle model using an absorbing-state Markov changepoint and Bayes factors (Chib 1995/1998). Found that a narrowing boom–recession gap (not just variance decline) is the dominant source of the Great Moderation: recession mean −1.025 → −0.195; boom mean 0.486 → 0.137. See Great Moderation and Kim-Nelson (1999b).
Related