Resolve Riffs Investment Podcast cover image

ReSolve Riffs with ReSolve’s own Dr Andrew Butler on Integrating Prediction with Optimization

Resolve Riffs Investment Podcast

00:00

Using Decision Tree Boosted Decision Tree Regression Models

In most cases, the goal of a decision tree is to minimize mean squared error. In this paper, you can actually reformulate the whole problem by using gradient boosting or decoupled approach. And it lends itself very nicely to different types of prediction modeling classes like decision boosting and gradient boosting models. The key findings were that for particular constraint sets, in particular when the mean variance portfolio was either unconstrained or has only linear equality constraints, then you get closed form optimal solutions for the integrated regression coefficients.

Play episode from 01:23:18
Transcript

The AI-powered Podcast Player

Save insights by tapping your headphones, chat with episodes, discover the best highlights - and more!
App store bannerPlay store banner
Get the app