Volatility Clustering, Risk-Return Relationship, and Asymmetric Adjustment in the Canadian Housing Market

Abstract

In this study, we apply a Lagrange multiplier (LM) test for the autoregressive conditional heteroscedasticity (ARCH) effects and an exponential generalized autoregressive conditional heteroscedasticity-in-mean (EGARCH-M) model to assess whether regional house prices in Canada exhibit financial characteristics similar to stock indices. Volatility clustering, positive risk-return relationships, and leverage effects are empirically shown to exist in the majority of provincial housing markets of Canada. These volatility behaviors, which reflect regional idiosyncrasies, are further found to differ across provinces. More densely populated provinces exhibit stronger volatility clustering of house prices. The existence of these volatility patterns similar to stock indices has important implications ranging from proper portfolio management to government policy.