What is the project about?

‍

‍

The main purpose of the project is to develop a technique, including the development of a prototype tool, that can help investors, including small wealth management offices, to take informed investment decisions that optimise a portfolio’s ESG impact, and, in particular, minimise the CO2 emissions of the underlying portfolio assets. ​

​

Whereas investors may use classic Markowitz mean-variance portfolio analyses to balance the tradeoffs between variance (risk) and the portfolio performance, ESG data is currently used mainly to filter out individual asset classes, rather than for optimising a portfolio’s ESG factors, given particular risk and performance levels, which may be computational expensive as ESG factors contribute with an additional dimension of computational needs.​

​

In this project, we aim at demonstrating that with sufficient computing power, accessed through massive-parallel programming technology targeting GPGPUs, it is possible to provide decision makers with valuable techniques for optimising and managing portfolio ESG factors, for realistic portfolios, while at the same managing the portfolio’s risk and performance through classic portfolio optimisation techniques. ​

​

The project is based on current research developed at the Department of Computer Science at Copenhagen University and done in collaboration with FinE and Invest Fondsmæglerselskab A/S.​

​

Host Institution(s):​

Copenhagen University​

Principal Investigators:​

Martin Elsman, Professor, Department of Computer Science​

Partners:​

FinE an Optimal Invest Fondsmæglerselskab A/S ​

Grant:​

200.000 DKK​

​

Material:​

Material will be posted upon completion of the project​

​

‍