In this paper we investigate the suggestion that the conventional stochastic disturbance term is not an essential feature of an economic model as it can be replaced by a suitable chaotic deterministic process. This prescription is clearly justified in the context of a simulation study when the deterministic process is carefully chosen for its pseudo-random properties. But this is not true of an arbtrarily selected chaotic process (such as the logistic map) whose aperiodic behaviour depends crucially on the strict continuity of its domain of definition. We also briefly examine the possibility of developing a test which discriminates between chaotic deterministic and truly stochastic processes. Two Appendices to the paper contain a Pascal program for computing the pattern of limit cycles for the rounded logistic map and a select bibliography of the economic applications of chaos theory.
[1]
George Sugihara,et al.
Nonlinear forecasting as a way of distinguishing chaos from measurement error in time series
,
1990,
Nature.
[2]
M. Casdagli.
Chaos and Deterministic Versus Stochastic Non‐Linear Modelling
,
1992
.
[3]
Robert M. May,et al.
Simple mathematical models with very complicated dynamics
,
1976,
Nature.
[4]
W. Brock.
Distinguishing random and deterministic systems: Abridged version
,
1986
.
[5]
M. Pohjola.
Stable, cyclic and chaotic growth: The dynamics of a discrete-time version of Goodwin's growth cycle model
,
1981
.
[6]
D. Y. Downham,et al.
Multiplicative congruential pseudo-random number generators
,
1967,
Comput. J..