Online Program Home
My Program

Abstract Details

Activity Number: 120 - SPEED: Nonparametric Statistics: Estimation, Testing, and Modeling
Type: Contributed
Date/Time: Monday, July 30, 2018 : 8:30 AM to 10:20 AM
Sponsor: Section on Nonparametric Statistics
Abstract #329942 Presentation
Title: Random Conditional Histogram Based Density Estimation with Applications in Probabilistic Forecasting
Author(s): Rui Li* and Howard D Bondell and Brian Reich
Companies: North Carolina State University and University of Melbourne and North Carolina State University
Keywords: nonparametric ; density estimation; neural network; probabilistic forecasting
Abstract:

While various machine learning methods have emerged as flexible nonparametric models for non-linear mean regression, many of them have not been extended to uncover full conditional distribution for uncertainty quantification. In applications like probabilistic forecasting, conditional density estimation is often required for future decision making. Here we present a framework where we transform conditional density estimation problem to randomized classification problem, which we can then harness the power of diverse classification algorithms. We evaluate our method in both simulation studies as well as a real data example concerning probabilistic forecasting of solar energy production.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2018 program