There are literally millions of quantitative investment strategies available - either published or unpublished, depending on their success in real market environments. Contemporary Deep Learning methods as well as the availability of alternative data allow for completely new structures of models to compute clever decisions for successful investment management, e.g. by using convolutional layers to examine images of various technical indicator charts or by integrating satellite images. Unfortunately it has also been shown that no technique will ever outperform the market over a reasonable long period of time. However, a clever semi-automatic selection of sophisticated and heterogeneous underlying investment models can facilitate investment magic. By using Interpretable Machine Learning techniques like standard CARTs the output of some meta-ensemble model can even be used to create a feedback loop with the respective investor. In this regard we present an application from the field of app-based Robo Advisory Portfolio Strategy Selection.
Ronald Hochreiter is Associate Professor of Finance at Webster Vienna Private University and Principal Investigator of various national and EU research projects at WU Vienna University of Economics and Business. He is senior partner at algorithmic.finance, a boutique consulting company tailoring bespoke Quantitative Asset Management Solutions for investment companies worldwide. Furthermore he is Associate Editor of the Journal 'Frontiers in Artificial Intelligence - AI in Finance' and President of the Academy of Data Science in Finance.