The predictability of rates of return on gold and silver are examined. Econometric tests do not reject the martingale hypothesis for either asset. This failure to reject is shown to be misleading. Correlation dimension estimates indicate a structure not captured by ARCH. The correlation dimension is between 6 and 7 while the Kolmogorov entropy is about 0.2 for both assets. The evidence is consistent with a nonlinear deterministic data generating process underlying the rates of return. The evidence is certainly not sufficient to rule out the possibility of some degree of randomness being present.
Bibliographical noteFunding Information:
Acknowledgement. We would like to thank Richard Arnott, William Brock, Roger Farmer, David Fowler, Clive Granger and Jose Scheinkman for valuable discussions. Lars Hansen's criticisms of an earlier draft were extremely helpful. The comments by the referees are appreciated. This research was partly supported by a grant from the Research Excellence Program of the University of Guelph. We are grateful to all of these individuals, but none of them are responsible for any remaining deficiencies in our work.