A0518
Title: The risk aversion coefficient in the tangency portfolio for large dimensions and a singular covariance matrix
Authors: Stanislas Muhinyuza - Linnaeus University (Sweden) [presenting]
Peter Karlsson - Linneaus University (Sweden)
Stepan Mazur - Orebro University (Sweden)
Abstract: The finite-sample distribution of the risk aversion coefficient of the tangency portfolio is derived in the form of a stochastic representation (SR). The focus is on the situation where both the population and the sample covariance matrices of asset returns are singular, particularly when the portfolio size is larger than the number of observations and the returns are identically and independently distributed. The derived SR is used to derive the moments of the risk aversion coefficient and to establish its high-dimensional approximation. Furthermore, through simulation, the good performance of the proposed high-dimensional approximation is documented.
