Detecting Asset Price Bubbles

This article reviews a new methodology for detecting asset price bubbles that is based on the statistical properties of the asset price process. The classical economic approach to test for bubbles involves constructing an equilibrium model and testing a joint hypothesis of no bubble and the model. In contrast, the stochastic process approach to test for bubbles involves constructing a stochastic process and estimating its parameters. The processes' parameter space is divided into two mutually exclusive and exhaustive subsets: One is consistent with a price bubble, but the other is not. As such, this alternative approach partially avoids the test of a joint hypothesis. The stochastic process testing methodology is an extension of the standard techniques used for estimating the underlying asset price processes for derivatives pricing.