Explainable Artificial Intelligence For Crypto Asset Allocation

Abstract

Many investors have been attracted by Crypto assets in the last fewyears. However, despite the possibility of gaining high returns,investorsbear high risks in crypto markets. To help investors and make the markets more reliable, Robot advisory services are rapidly expanding inthe field of crypto asset allocation. Robot advisors not only reducecosts but also improve the quality of the service by involving investorsand make the market more transparent. However, the reason behindthe given solutions is not clear and users face a black-box model thatis complex. The aim of this paper is to improve trustworthiness ofrobot advisors, to facilitate their adoption. For this purpose, we apply Shapley values to the predictions generated by a machine learningmodel based on the results of a dynamic Markowitz portfolio optimiza-tion model and provide explanations for what is behind the selectedportfolio weights.

Keywords: Machine learning, Shapley values, Robo-advisory