CMStatistics 2018: Start Registration
View Submission - CFE
Title: Testing for parametric orderings efficiency Authors:  Matteo Malavasi - Macquarie University (Australia) [presenting]
Sergio Ortobelli - Universita degli Studi di Bergamo (Italy)
Nikolas Topaloglou - Athens University of Economics and Business Research Center (Greece)
Abstract: Semi-parametric tests to evaluate the efficiency of a benchmark portfolio with respect to different stochastic orderings are developed and empirically compared. Firstly, we classify investors' choices when returns depend on a finite number of parameters: a reward measure, a risk measure and other parameters. We extend stochastic dominance theory under minimal assumptions on reward and risk measures. We prove that, when choices depend on a finite number of parameters and, when the reward measure is isotonic with investors' preference, agents behave as non satiable and risk averse, when the reward measure is lower than the mean, and behave as non satiable and risk seeker when the reward measure is greater than the mean. Then, we introduce a new stochastic ordering that is consistent with the choices of a non satiable, nor risk averse nor risk seeker investor. Secondly, we propose a methodology to semi-parametric tests for the efficiency of a portfolio, when return distribution is uniquely identified by four parameters, using estimation function theory. Finally, we empirically test whether the Fama and French market portfolio, as well as the NYSE and the Nasdaq indexes are efficient with respect to alternative stochastic orderings.