Kalman filtering estimation of unobserved rational expectations with an application to the German hyperinflation

Abstract The assumption that rational expectations always lie on a convergent path is subject to an empirical test using the German hyperinflation data. The estimation technique employs a Kalman filtering algorithm. After presenting a brief background for the convergent expectations problem and a derivation of the various model specifications, a generalized expectations model and its attendant Kalman filtering estimation technique are discussed. Additional estimation details and empirical results are then presented. Based on an assumption of normally distributed errors, the null hypothesis of convergent paths is rejected in all situations involving a deterministic specification of the evolution of the unobserved parameter which characterizes the convergent path. The same null hypothesis is rejected in four of the six cases corresponding to a stochastic specification of the evolution of the unobserved parameter which characterizes the convergent path. A discussion of these findings, their economic significance, and suggestions for further research concludes the paper.

[1]  Box Ge,et al.  Time series analysis: forecasting and control rev. ed. , 1976 .

[2]  Thomas J. Sargent,et al.  Rational Expectations and the Dynamics of Hyperinflation , 1973 .

[3]  R. Bucy,et al.  Filtering for stochastic processes with applications to guidance , 1968 .

[4]  P. Caines,et al.  Linear system identification from non-stationary cross-sectional data , 1978, 1978 IEEE Conference on Decision and Control including the 17th Symposium on Adaptive Processes.

[5]  L. Ljung.  Convergence analysis of parametric identification methods , 1978 .

[6]  Robert P. Flood,et al.  Market Fundamentals versus Price-Level Bubbles: The First Tests , 1980, Journal of Political Economy.

[7]  T. Sargent,et al.  FORMULATING AND ESTIMATING DYNAMIC LINEAR RATIONAL EXPECTATIONS MODELS , 1980 .

[8]  J. Muth.  Rational Expectations and the Theory of Price Movements , 1961 .

[9]  Edwin Burmeister,et al.  On Some Conceptual Issues in Rational Expectations Modeling , 1980 .

[10]  Jerry R. Green,et al.  Variance-Minimizing Monetary Policies with Lagged Price Adjustment and Rational , 1983 .

[11]  F. Black.  Uniqueness of the price level in monetary growth models with rational expectations , 1974 .

[12]  A. Jazwinski.  Stochastic Processes and Filtering Theory , 1970 .

[13]  Thomas F. Cooley,et al.  A Note on Optimal Smoothing for Time Varying Coefficient Problems , 1976 .

[14]  B. Anderson,et al.  Optimal Filtering , 1979, IEEE Transactions on Systems, Man, and Cybernetics.

[15]  P. Caines.  Prediction error identification methods for stationary stochastic processes , 1976 .

[16]  Gwilym M. Jenkins,et al.  Time series analysis, forecasting and control , 1972 .

[17]  Arthur E. Bryson,et al.  Applied Optimal Control , 1969 .

[18]  L. Godfrey.  TESTING AGAINST GENERAL AUTOREGRESSIVE AND MOVING AVERAGE ERROR MODELS WHEN THE REGRESSORS INCLUDE LAGGED DEPENDENT VARIABLES , 1978 .

[19]  P. Caines,et al.  Asymptotic normality of prediction error estimators for approximate system models , 1980 .

[20]  P. Caines,et al.  Linear system identification from nonstationary cross-sectional data , 1979 .

[21]  B. Mccallum.  Rational Expectations and Macroeconomic Stabilization Policy: An Overview , 1980 .

[22]  K. Wall.  Generalized expectations modeling in macroeconometrics , 1980 .

[23]  Robert P. Flood,et al.  On the equivalence of solutions in rational expectations models , 1983 .