In this first episode of the Curious Quant podcast series, I sit down with Jim Creighton from Creighton AI. We talk about everything from the history of quantitative investment and the myths of factor investing to how machine learning can help solve difficult problems in financial markets. A former CIO of Barclays Global Investors, Creighton has spent his career applying mathematical principles to financial concepts. He delivers some excellent observations.

The future of asset management:

“I would make a passionate argument that the potential for AI in asset management is so big, that if we could fast forward ten or fifteen years…it will be the dominant methodology.”

Inherent challenges in financial markets:

“We know the world is not linear, we know that we can’t write down the equation that links returns and outcomes, we can’t even write the probability density function that describes financial returns.”

Predicting the next move:

“When you focus solely on fundamental factors, whether they are valuation factors or earnings factors, you are pretty much ignoring the fact that what you are trying to predict is the next movement in a time series. What you are assuming is that these types of factors will predict the next movement. The route I have taken over the last decade is to focus on the time series itself, the structural property of the times series move, what information is contained within the times series which will help you predict the next probable move.”

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