Description: Developing reliable and intuitive interpretation is essential for the application of machine learning to investing. This presentation discussed a framework for decomposing any machine learning model into linear, nonlinear, and interaction effects that drive both prediction and performance.

Session 1: Portfolio for the Future
Description: How can investment professionals meet their clients' needs under ever-changing economic and market conditions? We call future-proofed portfolios the Portfolio for the Future™. The five distinct marks of the Portfolio for the Future™ are the defining features of all-weather portfolios that meet the challenges of the global economy and capital markets. The 5 marks are – Broadly Diversified, Less Liquid, Rooted in a Fiduciary Mindset, Actively Engaged and Dependent on Operational Alpha.

Description: Artificial Intelligence (AI) has been a great buzzword in scientific and business communities and even the popular press. This webinar was designed to explore the technology underlying AI, known as “unsupervised learning.” Unsupervised learning allows us to hear the data speak for itself, reducing reliance on researchers’ potentially biased frameworks and sometimes discovering unexpected factors. In the course of the webinar, we considered AI applications in Finance and contrasted those with traditional machine learning (ML) models.

Description: Asset managers have struggled to apply machine learning to generate trading signals successfully. Instead, applying machine learning to determine the probability of profit of existing trading signals is very fruitful - this is Corrective AI. Taking this one step further, we developed Conditional Portfolio Optimization, a portfolio optimization technique that adapts to market regimes via machine learning. Applications on portfolios in vastly different markets suggest that CPO can outperform traditional optimization methods under varying market regimes.