Deterministic chaos and fractal attractors as tools for nonparametric dynamical econometric inference: With an application to the Divisia monetary aggregates (Q1110432)

A challenging approach to monetary policy is controlling monetary aggregates by the policy makers. A new line of research to solve this problem has been offered recently by the divisia monetary aggregates theory. In the present paper the authors combine this theory with the deterministic chaos and strange attractor theory to draw nonparametric econometric inference. To apply this statistical tool high-quality data with a low noise-to-signal ratio are required. The authors find that for the demand side the detrended aggregates M2 and M3 possess characteristics of mathematical chaos. Combining these results with low-dimensional state space models, they conclude that either the noise-to-signal ratio is too high for alternative monetary aggregates or that the data-quality is high, but the alternative aggregates cannot be controlled by a small number of control variables. The paper is divided into two parts. The first part reviews briefly mathematical chaos and strange attractor theory; the second part contains the application.