Robo-advising under rare disasters

Published: 17 March 2022

8 April. A case of Covid-19 onset

robot hand pointing to graph

A case of COVID-19 onset

Friday 8 April 2022, 2:00pm - 3:00pm 
Online

The University of Glasgow Adam Smith Business School's Finance cluster are delighted to be hosting a webinar on Friday 8 April 2022 with Professor Cathy Yi-Hsuan Chen. This webinar is free to attend and will include a presentation followed by a Q&A session. 

Register on Eventbrite now

The rise of Robo-advisors and their growth is unprecedented. Robo-advisors, compared to the preceding period, stumble during the COVID-19 pandemic. Such a less satisfying performance is due to the presence of rare disasters for their low probability of occurrence and a huge impact on the markets. To improve the performance and learning efficiency of Robo-advising, we develop a novel learning framework to tackle rare disaster events. It brings the importance of sampling technique into a reinforcement learning algorithm. Instead of sampling the transition probability from the ground-truth probability distribution, one sample from the proposal distribution, in which the event of interest occurs more frequently. The proposed algorithm is validated with the 2008 financial crisis and the COVID-19 pandemic data, showing superior performance over the competing algorithms. The estimated quarterly return using the optimal policy is 0.069%, compared to -3.56% and -14.55% generated by the benchmark policy and the average quarterly return of existing Robo-advisor portfolios, respectively. Such a superiority is attributed to intended learning toward disaster events. 

Cathy Yi-Hsuan Chen is a professor of Corporate Finance and Banking at the University of Glasgow, Adam Smith Business School. She has been the Mercator Fellow of the International Research Training Group 1792 in Germany since 2018. Her publications are deep in theory and broad in scope and have had a strong impact on social science research. She has specialized herself in network analysis, Blockchain economy, FinTech era, and data analytics. She is a fellow in Blockchain Research Center, a think-tank and research community established by the Universität Zürich and the Humboldt-Universität zu Berlin. She is one of the PIs on an interdisciplinary EPSRC-funded project on 'Privacy-preserving and Secure Data Sharing and Trading Ecosystem for Distributed Wireless IoT Networks’. She represents the UK government, Department of International Trade, for reviewing FinTech awards 2019. She has been nominated as a Management Committee substitute representing the UK for the EU FinTech Action.


Further information: business-events@glasgow.ac.uk

First published: 17 March 2022

<< 2022