Estimates of the Bias of Lagged Dependent Variable Coefficient Estimates in Macroeconomic Equations

A stochastic-simulation method is proposed in this paper for obtaining median unbiased estimates of lagged dependent variable coefficients in macroeconomic models. Estimated biases for 13 equations of a macroeconomic model are computed. These biases are on average somewhat smaller in absolute bias than would be expected from Andrews' exact results for an equation with only a constant term, time trend, and lagged dependent variable, although they are larger than would be expected from Hurwicz's original estimates. In a practical sense the estimated biases are not very large because they have little effect on the overall predictive accuracy of the model and on its multiplier properties.