In order to supply additional empirical evidence of the effect of wealth on relative risk aversion, this study investigates households' demand for risky assets, using analysis of covariance techniques applied to the asset holdings of Canadian individual households. The extent and pattern of life‐cycle effects are also examined. Results generally point to decreasing relative risk aversion when housing is either excluded from the definition of wealth or treated as a riskless asset. The investor's life‐cycle plays a prominent role in portfolio selection behavior, with risk aversion increasing uniformly with age. Tax differentials do not seem to be an important element in investment decisions with respect to risk. When the sample and wealth definitions are censored in order to approximate those of previous empirical studies, their findings on relative risk aversion are generally corroborated.