Van Appel, VaughanMare, Eben2019-10-242019-10-242018-11-04Van Appel, V. & Mare, E. 2018, 'The Ross recovery theorem with a regularised multivariate Markov chain', Orion, vol. 24, no. 2, pp. 133-155.0529-191X (print)2224-0004 (online)10.5784/34-2-594http://hdl.handle.net/2263/71974Recently, Ross derived a theorem, namely the “Recovery Theorem”, that allows for the recovery of the pricing kernel and real-world asset distribution, under particular assumptions, from a forward-looking risk neutral distribution. However, recovering the real-world distribution involves solving two ill-posed problems. In this paper, we introduce and test the accuracy of a regularised multivariate mixture distribution to recover the real-world distribution. In addition, we show that this method improves the estimation accuracy of the real-world distribution. Furthermore, we carry out an empirical study, using weekly South African Top40 option trade data, to show that the recovered distribution is in line with economic theory.enOperations Research Society of South AfricaReal-world probabilitiesRoss recovery theoremMultivariate Markov chainArticle