AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
The Challenges With Optimization
We've been leaning heavily on concepts like bagging, unsupervised learning. You're creating clusters of similar assets and then you use those clusters to create a structured re-sampling process. And it's funny because the idea of shrinkage is linked to re-Sampling, is linked to regularization and an optimization process itself is linked to constraints. The long only constraint does a miraculous job of producing much more stable portfolios.