Identifying reference companies using the book-to-market ratio: a minimum spanning tree approach

There is substantial evidence to suggest that the book-to-market (BM) ratio is an important factor in explaining stock market returns. Its role has proved difficult to isolate, however, due to statistical problems in its construction and to its observational equivalence to a number of risk and behavioural explanations. In addition, now widely recognised complex behaviour in financial markets has called into question modelling approaches that are limited in their ability to uncover relationships that are possibly masked during financial crises, for example. As one response, our research explores the value of a newly applied technique which examines the topological properties of minimum spanning trees as applied to both the BM ratio and market returns. Our intention is to identify and report investment signals as determined by the BM ratio and to assess the relationships of these signals to returns outcomes. The approach enables highly nonlinear behaviour to be addressed and the relationships we set out to capture to be reported in novel ways. We motivate and evidence a previously unreported role for BM as an investment signal which is effective over varying stock market conditions, including the financial crisis that began in 2008.