The predictive ability of value at risk (var) for equity returns: evidence from New Zealand

Mark Brimble
Tim Lukin

Abstract: The accurate and timely estimation of investment risk has been the focus of much attention in the applied finance literature. With the increasing sophistication of the capital markets and the proliferation of derivative instruments, traditional risk measurement techniques have struggled to meet the needs of investors. A relatively new risk measure that has the ability to capture a portfolio’s exposure to market movements is Value at Risk (VaR). It is a single summary measure of portfolio risk that estimates the maximum potential loss that a portfolio can endure, given a specified time period and confidence interval. To date, there are only a few papers that empirically examine the usefulness of VaR with little evidence on small economies. This study proposes to fill this gap in the literature by examining the predictive ability of VaR in the New Zealand equity market. VaRs are calculated for a sample of 100 listed New Zealand companies over weekly, monthly, quarterly and half yearly holding periods. Portfolios are then constructed based on the magnitude of the VaR estimates and compared based on the average abnormal returns to the portfolios. The results evidence a positive relationship between VaR and portfolio returns in the short term, which conforms with traditional portfolio theory. Furthermore, the length of the holding period is an important determinant in the relationship between VaR and equity returns. This finding provides initial evidence of the usefulness of VaR as a predictive measure of equity performance in small economies.