Affordable Access

Annals of Economic and Social Measurement, Volume 5, number 2

  • Economics
  • Mathematics


NEIGHBORING STOCHASTIC CONTROL OF AN ECONOMETRIC MODEL This PDF is a selection from an out-of-print volume from the National Bureau of Economic Research Volume Title: Annals of Economic and Social Measurement, Volume 5, number 2 Volume Author/Editor: NBER Volume Publisher: Volume URL: Publication Date: April 1976 Chapter Title: Neighboring Stochastic Control of an Econometric Model Chapter Author: Peter Walsh, J. B. Cruz, Jr. Chapter URL: Chapter pages in book: (p. 211 - 221) An,ials of Ecotwmic and Social Measurement, 5/2, 1q76 NEIGHBORiNG STOCI IASTIC CONTROL OF AN ECONOMETRIC MODEL* RY PETER WALSH AND J. B. CRUZ, JR. In this paper, an econometric model with parameter uncertainty considered by Kendrick an,I Majors, which extends the deterministic linear econometric model used by Pindyck. is modified further to account for additive errors in the strucuiral equations and additive observa lion errors in variables. The use of a linearized neighboring optimal stochastic control with a Kalmnan filter is investigated. Simulation results indicate that when a Kalman filter is used to improve the estimates of the state variables and parameters. deviations from the desired paths tend to be attenuated. 1. INTRODUCTION In recent years there have been many applications of modern control theory to economic stabilization and planning, such as the work done by Chow [4, 5], Friedman [6], Livesey [10], Pindyck [15, 16], and Sengupta [18]. Pindyck [15] used a 28 state variable linearized model, assumed to be deterministic, and a quadratic tracking approximation for the criterion function. The validity of a linearized model has been discussed by Pindyck [15, 16] and the economic justification of the quadratic criterion approximation has been discussed by Theil [19]. An approach to the problem which takes parameter variations into account has been done by Hendrick and Majors [8] utilizing Pin

There are no comments yet on this publication. Be the first to share your thoughts.


Seen <100 times