No label defined (Q4495096)

This paper is devoted to a survey of the classical results of static mean-variance analysis and extension of these ideas to continuous-time dynamic models. Consider an investor who has a vector of \(N\) assets with returns \(S=(S_1,\ldots,S_{N})'\), where these are jointly normally distributed with the vector of means \(\mu=(\mu_1,\ldots, \mu_{N})'\) and the symmetric, positive definite covariance matrix \(\Gamma\). Let us consider a money market with interest rate \(r\), \(0\leq r<m_0\). Assume that \(\mu\) and \(e=(1,\ldots,1)\) are linearly independent. The problem of choosing a best portfolio for a particular agent is separated into two parts. First the agent should determine the market portfolio, a portfolio whose composition depends only on the asset parameters, not on the preferences of the individual agent. Secondly, the agent should allocate her wealth between the market portfolio and the money market according to her attitude toward risk. The Capital Asset Pricing Theorem is presented also both in static and continuous-time dynamic models. For the entire collection see [Zbl 0937.00058].