Speaker: Chris Adcock, Professor of Quantitative Finance, SOAS University of London
Chair: Victor Murinde, AXA Professor in Global Finance, SOAS University of London

Abstract

It is widely believed that “all correlations go to one in a crisis". Under the market model, unit correlations can only arise if residual risks vanish and asset returns are determined only by the return on the market portfolio. This paper investigates the changes in correlation that occur when the conditional distribution of asset returns given a market crash is used. Assuming a multivariate normal distribution for asset returns under this conditioning: (i) returns have a skewed distribution when there is a market crash, but the market is the sole source of skewness; (ii) the betas remain unchanged but; (iii) that specific changes are required to the parameters of the underlying multivariate normal distribution of returns if unit correlations are to be observed. It is also shown that whether or not there are changes in the underlying parameters the market return becomes deterministic as the crash magnitude without limit. If returns follow a multivariate Student distribution, results (i) to (iii) hold essentially unchanged. In sharp contrast, the market return distribution has a standard deviation that increases with the magnitude of the crash. The paper also considers linear factor models of the Fama and French type. Under both normal and Student distributions unit correlations will only arise if the factors themselves are perfectly correlated. Under both the distributions considered in this paper, unit correlations require complex changes to model parameters, which offers some support for the mixed findings reported in the literature about correlations at times of market crashes.

 A sandwich lunch will be served.

The seminars are sponsored by grants from DFID and ESRC [ESRC Ref: ES/N013344/2], ESRC and NSFC [ESRC Ref: ES/P005241/1] and AXA Research Fund