Historical performance of robot advisors – an historical backtest and comparative study

• Autor(a) principal: Nowottny, Andreas
• Data de Publicação: 2019
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10362/70415
• topic: Robo advisor; Portfolio management; Markets; Performance; Backtest; Domínio/Área Científica::Ciências Sociais::Economia e Gestão

What do portfolios offered by robo advisors look like in practice? And how would these portfolios have performed over the past 17.5 years, considering the dot-com bubble as well as the 2008/2009 global financial crisis? Current robo advisory portfolios from mainly German fintechs are analysed and backtested, utilizing an ETF-replication approach in which current ETF fees are deducted from historical, total return indices to mimic the performance of these financial products over the analysed time period. The performance statistics and factor loadings are then analysed and compared. The obtained results show that asset allocations in general do not differ greatly for similar risk-attitudes. However, the influence of fees as well as stronger loadings on the factors of size and value can have a substantial impact on the overall portfolio performance after controlling for different risk levels. Concluding, the analysed robo advisors do not outperform a simple, self-managed ETF portfolio, but do add value in terms of possible factor exposure, the quantification of risk-attitudes as well as the full automation of the asset allocation and investment process. Overall, this is a practical work with a strong focus on the implemented backtest, hence consulting the attached Excel model is strongly advised.