Are efficient markets really efficient?: can financial econometric tests convince machine learning pepople?

摘要

Using Qninlan's Cubist, this paper examines whether there is a consistent interpretation of the efficient market hypthesis between financial econometrics and machine learning. In particular, we asked whether machine learning can be useful only in the case that the market is not efficient. Based on the forecasting performance of Cubist in our artificial returns, some evidences seems to support this consistent interpretation. However, there are a few cases that Cubist can bat the random walk even though the series is independent. As a result, we do not consider the evidence is strong enough to convince machine learning to give up when the efficient market hypothesis is sustained.