In this episode of the Curious Quant podcast series, I sit down with Linda Gruendken from GAM Systematic, CANTAB, where we look at the uncertainty in markets and how AI can bring a level of truth to the analysis. Reflecting on her academic career and her experiences in devising rules for machine learning investment processes, we discuss the future for credit market modelling. She makes some excellent observations.

On AI programmes:

“The process itself does not need us, its fully independent.”

On minimising human bias:

“We encode rules for investment, we do a lot of testing, but then we don’t allow ourselves to interfere because we want to eliminate the human bias.”

On the biggest challenge:

“We know we can build many indicators, we can test many variations of similar indicators, but the problem we have in finance is that we have a lot of noise in market, we have to avoid overfitting.”

You can download the podcast from the Apple and Spotify podcast players or listen below:

Dr Michael G. Kollo is a seasoned investment professional with a strong academic background, and a deep passion for the pragmatic discussion and application of quantitative models (like AI, ML) to solve problems for different problem spaces. He gained his PhD in Finance from the London School of Economics, and has lectured at the London School of Economics, Imperial College and at the University of New South Wales on the topic of quantitative finance. Dr Kollo’s global industry career spans from London UK, where he led created models and led quantitative research teams at Blackrock, Fidelity, Axa Rosenberg to more recently moving to Australia where he established the quantitative team for HESTA a $50bn superannuation industry fund in Australia.