An ensemble method: Case-based reasoning and the inverse problems in investigating financial bubbles

Abstract

This paper presents an ensemble approach and model; IPCBR, that leverages the capabilities of Case based Reasoning (CBR) and Inverse Problem Techniques (IPTs) to describe and model abnormal stock market fluctuations (often associated with asset bubbles) in time series datasets from historical stock market prices. The framework proposes to use a rich set of past observations and geometric pattern description and then applies a CBR to formulate the forward problem; Inverse Problem formulation is then applied to identify a set of parameters that can statistically be associated with the occurrence of the observed patterns. The technique brings a novel perspective to the problem of asset bubbles predictability. Conventional research practice uses traditional forward approaches to predict abnormal fluctuations in financial time series; conversely, this work proposes a formative strategy aimed to determine the causes of behaviour, rather than predict future time series points. This suggests a deviation from the existing research trend.