Should Macroeconomic Policy Makers Consider Parameter Covariances?

Authors:
Hans M. Amman;David A. Kendrick
Affiliations:
Department of Economics, University of Amsterdam, Roetersstraat 11, 1018 WB Amsterdam, The Netherlands E-mail: amman@fee.uva.nl;Department of Economics, University of Texas, Austin, Texas 78712, U.S.A. E-mail: kendrick@eco.utexas.edu
Venue:
Computational Economics
Year:
1999

Citing 0
Cited 9

The Timing of Uncertainty and the Intensity of Policy

Computational Economics
Uncertainty, Political Preferences, and Stabilization: Stochastic ControlUsing Dynamic CGE Models

Computational Economics
Parameter Uncertainty and Policy Intensity: Some Extensions and Suggestions for Further Work

Computational Economics
A Classification System for Economic Stochastic Control Models

Computational Economics
Stochastic Ceteris Paribus Simulations

Computational Economics
Optimal Policy Response with Control Parameter and Intercept Covariance

Computational Economics
Solving the Beck and Wieland model with optimal experimentation in DualPC

Automatica (Journal of IFAC)
Optimal Deterministic and Stochastic Macroeconomic Policies for Slovenia: An Application of the OPTCON Algorithm

Computational Economics
Optimal control of nonlinear dynamic econometric models: An algorithm and an application

Computational Statistics & Data Analysis

Quantified Score

Hi-index	0.02

Visualization

Abstract

Many macroeconomic policy exercises consider the mean values of parameterestimates but do not use the variances and covariances. One can argue that theuncertainty of these parameter estimates is sufficiently small that it cansafely be ignored. Or one can take the position that this kind of uncertaintycannot be avoided no matter what one does. Thus it is just as well to ignoreit while making policy decisions.In this paper we address both of these positions in the presence of learningand find that they are unconvincing. To the contrary, we find evidence thatthe potential damage from ignoring the variances and covariances of theparameter estimates is substantial and that taking them into account canimprove matters.