Interval Prediction for Spatial Data Using Deep Learning Models
Statistical analysis for the purpose of prediction is preferably accompanied by uncertainty quantification, often in the form of prediction intervals. Deep learning approaches have been extensively shown to provide accurate point predictions in many applications.
Overview
Abstract
Statistical analysis for the purpose of prediction is preferably accompanied by uncertainty quantification, often in the form of prediction intervals. Deep learning approaches have been extensively shown to provide accurate point predictions in many applications. However, the generation of prediction intervals within conventional deep-learning models calls for diligent adaptation, development, testing, and implementation. To this end, we propose a novel deep learning model which provides both accurate point predictions and prediction intervals, through empirical score minimization of proper scoring rules for interval forecasts. In simulation studies and real data applications, we demonstrate the efficacy of the novel deep learning model against traditionally used methods.
Brief Biography
Ghulam Qadir is originally from India. He studied Statistics in Bachelors, Masters and Completed Ph.D. in Statistics from King Abdullah University of Science and Technology (KAUST). His research interest: Spatial and Spatio-temporal Statistics, Deep Learning, Spatial Deep Learning. He is currently a Postdoc in the Computational Statistics group at Heidelberg Institute for Theoretical Studies, Germany. Ghulam is Recipient of KAUST's Alkindi Student Research and Alkindi Top Qual Award.