url
23
May
2014

SIMPLE MONETARY POLICY RULES UNDER MODEL UNCERTAINTY: Comparison of Basic Model Properties 7

Table 2 indicates the asymptotic values of 6, 0, and ั„ for each model computed using the unconditional moments of each model and the estimated coefficients and standard errors obtained from fitting equation (2) using U.S. data over the sample period 1966Q1 – 1995Q4. Although the MSR model and TAYMCM have roughly similar output autocorrelograms, the table shows that these models actually imply very different behavior for the output gap. In particular, the output gap in TAYMCM error-corrects to zero rapidly and displays essentially no accelerator effect, whereas the output gap in the MSR model error-corrects more gradually and displays a strong accelerator effect. The FM and FRB models imply roughly similarly low rates of error-correction, while the accelerator effect in FM is nearly twice as strong as in the FRB model. Finally, the coefficients on the real short-term interest rate are similar across the four models, with FM displaying the least real interest rate sensitivity of output.

Models U.S. Data
Coefficient FM MSR TAYMCM FRB 1966:1-1995:4
6 -0.04 -0.17 -0.31 -0.05 -0.10(0.03)
0 0.49 0.45 -0.04 0.26 0.29(0.08)
ะค -0.03 -0.05 -0.06 -0.04 -0.08(0.03)

Policy Frontiers for Simple and Complicated Rules
This section outlines the objective function and constraints used in determining the inflation-output volatility frontier of each model for a given specification of the policy rule.

Then we analyze the properties of these frontiers for several alternative specifications: rules in which the federal funds rate responds to only three variables (the current output gap, a moving average of the inflation rate, and the lagged funds rate), more complicated rules that incorporate a larger number of observed state variables, and rules that incorporate model-based forecasts of the output gap and inflation rate. Finally, we consider the extent to which these results are sensitive to a one-quarter delay in information.

Table 2 indicates the asymptotic values of 6, 0, and ั„ for each model computed using the unconditional moments of each model and the estimated coefficients and standard errors obtained from fitting equation (2) using U.S. data over the sample period 1966Q1 – 1995Q4. Although the MSR model and TAYMCM have roughly similar output autocorrelograms, the table shows that these models actually imply very different behavior for the output gap. In particular, the output gap in TAYMCM error-corrects to zero rapidly and displays essentially no accelerator effect, whereas the output gap in the MSR model error-corrects more gradually and displays a strong accelerator effect. The FM and FRB models imply roughly similarly low rates of error-correction, while the accelerator effect

About The Author

Kevin J. Brandon

Home | Site Map | Contacts

Copyright © 2013 - 2019 Investment And Finance Online. All rights reserved