A Numerical Procedure to Estimate Real Business Cycle Models Using Simulated Annealing
This paper presents a numerical procedure to estimate a stochastic growth model of real business cycle type where the decision rules are not analytically solvable. The statistical estimation of this type of models faces the difficulty that the relationship is not explicit between the estimated parameters and the objective function for the estimation (for example, the distance function in GMM estimation). Furthermore, multiple local optima may exist since the objective function is often nonlinear in parameters. To circumvent these problems, we introduce a global optimization algorithm, the simulated annealing, that searches the parameter space recursively for the global optimum. When this algorithm is employed, along with an appropriate approximation method of dynamic programming to numerically compute the decision rules, the estimation turns out to be efficient. The estimation of structural parameters with statistical methods appears to us a necessary step to empirically evaluate RBC models.
KeywordsSimulated Annealing Stochastic Dynamic Programming Business Cycle Theory Real Business Cycle Macroeconomic Time Series
Unable to display preview. Download preview PDF.
- Chow, G.C., 1991, ‘Dynamic optimization without dynamic programming’, Econometric Research Program, Research Memorandum, No. 361, Princeton: Princeton University.Google Scholar
- Chow, G.C., 1993a, `Statistical estimation and testing of a real business cycle model’, Econometric Research Program, Research Memorandum, No. 365, Princeton: Princeton University.Google Scholar
- Christiano, L.J., 1987, `Dynamic properties of two approximate solutions to a particular growth model’, Research Department Working Papers, No. 338, Federal Reserve Bank of Minneapolis.Google Scholar
- Christiano, L.J. and M. Eichenbaum, 1992, ‘Current real business cycle theories and aggregate labor market fluctuation’, American Economic Review 82, 431–472.Google Scholar
- Goffe, W.L., G. Ferrier, and J. Rogers, 1991, ‘Global optimization of statistical functions’, in H.M. Amman, D.A. Belsley, and L.F. Pau (Eds), Computational Economics and Econometrics, Vol. 1, Dordrecht: Kluwer.Google Scholar
- Hansen, L.P. and K.J. Singleton, 1982, ‘Generalized instrument variables estimation of nonlinear rational expectations models’, Econometrica 50, 1268–1286.Google Scholar
- Semmler, W. and G. Gong, 1994, `Estimating and evaluating equilibrium business cycle models’, Working Papers Series, No. 56, Department of Economics, New School for Social Research.Google Scholar
- Semmler, W. and G. Gong, 1996a, ‘Estimating parameters of real business cycle models’, Journal of Economic Behavior and Organization, 30.Google Scholar
- Semmler, W. and G. Gong, 1996b, ‘Estimating stochastic growth models: A comparison of different data sets’, New School for Social Research, Memeo.Google Scholar
- Taylor, J.B. and H. Uhlig, 1990, ‘Solving nonlinear stochastic growth models: A comparison of alternative solution methods’, Journal of Business and Economic Statistics 8 (1), 1–17.Google Scholar