TIME-VARYING PARAMETERS AND NONCONVERGENCE TO RATIONAL-EXPECTATIONS UNDER LEAST-SQUARES LEARNING

被引：11

作者：

BULLARD, J

机构：

[1] Federal Reserve Bank of St. Louis, St. Louis, MO

来源：

ECONOMICS LETTERS | 1992年 / 40卷 / 02期

关键词：

D O I：

10.1016/0165-1765(92)90216-L

中图分类号：

F [经济];

学科分类号：

02 ;

摘要：

Least squares learning does not represent fully rational behavior because agents use inference techniques based on a parameter constancy assumption to learn about systems with time-varying parameters. When agents take time-varying parameters into ar-count, convergence results are lost.

引用

页码：159 / 166

页数：8

共 13 条

[1] INTRODUCTION TO THE STABILITY OF RATIONAL-EXPECTATIONS EQUILIBRIUM
BLUME, LE
BRAY, MM
EASLEY, D
[J]. JOURNAL OF ECONOMIC THEORY, 1982, 26 (02) : 313 - 317
[2] DUFFY J, 1991, UNPUT LEARNING RATIO
[3] FRYDMAN R, 1982, AM ECON REV, V72, P652
[4] HARVEY AC, 1982, GAMES EC DYNAMICS TI, P306
[5] ANALYSIS OF RECURSIVE STOCHASTIC ALGORITHMS
LJUNG, L
[J]. IEEE TRANSACTIONS ON AUTOMATIC CONTROL, 1977, 22 (04) : 551 - 575
[6] Ljung L, 1983, THEORY PRACTICE RECU
[7] MARCET A, 1988, AM ECON REV, V78, P168
[8] CONVERGENCE OF LEAST-SQUARES LEARNING-MECHANISMS IN SELF-REFERENTIAL LINEAR STOCHASTIC-MODELS
MARCET, A
SARGENT, TJ
[J]. JOURNAL OF ECONOMIC THEORY, 1989, 48 (02) : 337 - 368
[9] CONVERGENCE OF LEAST-SQUARES LEARNING IN ENVIRONMENTS WITH HIDDEN STATE VARIABLES AND PRIVATE INFORMATION
MARCET, A
SARGENT, TJ
[J]. JOURNAL OF POLITICAL ECONOMY, 1989, 97 (06) : 1306 - 1322
[10] A TIME-VARYING MODEL OF RATIONAL LEARNING
MARGARITIS, D
[J]. ECONOMICS LETTERS, 1990, 33 (04) : 309 - 314

← 1 2 →