The DSI forms a nexus for teaching and learning in data science, and for coordinating degree programs in data science that span departments. The Department of Mathematics is one example of these leading centres in the social sciences and provides the subject of our next data science spotlight.
Weiguan Wang is a recent PhD graduate from the Department of Mathematics with research interests including Machine Learning with applications to Quantitative Finance.
Weiguan's PhD thesis was titled 'Statistical Hedging with Neural Networks' and was supervised by Dr Arne Lokka and DSI Affiliate Professor Johannes Ruf.
Johannes has also collaborated on research with Weiguan and when asked about him explained that "what I am personally most impressed with about Weiguan is his strong scholarly attitude. He is always trying to understand what is going on and is very honest about his results." Information on some of their collaborations can be found below.
'Neural networks for option pricing and hedging'
This literature review was published in the Journal of Computational Finance and adds to the field of research through comparing papers in terms of input features, output variables, performance measures, and more.
'Hedging with linear regressions and neural networks'
In this article Weiguan and Johannes explore neural networks as nonparametric estimation tools for the hedging of options. The authors design a network, named HedgeNet, that directly outputs a hedging strategy that can be applied to end-of-day and tick prices of S&P 500 and Euro Stoxx 50 options. This article is forthcoming in the Journal of Business & Economic Statistics and was presented by Weiguan at the LSE Financial Mathematics Reading Group.
'Hedging with machine learning methods'
Weiguan has also published a blog on this joint work that provides an accesible explantion of the research. In this blog, Weiguan explores how machine learning methods are able to outperform the traditional method, and justify such an advantage with a well-known financial market fact.
Weiguan can be contacted via email and followed on GitHub.