• Corpus ID: 51862263


  author={Timothy DelSole and Claire Monteleoni and Scott McQuade and Michael K. Tippett and Kathleen Pegion and Jagadish Shukla},
A machine learning algorithm for combining predictions is applied to seasonal predictions of the NINO3.4 index from six coupled atmosphere-ocean models. The algorithm adaptively tracks a dynamic sequence of “best experts” and produces a probability that a particular expert is best. Averaging based on this probability effectively yields a multi-model prediction. The algorithm gives seasonal predictions that are more skillful than any individual model and better than the multi-model mean. 

Figures from this paper

A new approach to extended‐range multimodel forecasting: Sequential learning algorithms
Multimodel combinations are a well‐established methodology in weather and climate prediction and their benefits have been widely discussed in the literature. Typical approaches involve combining the
Big Data Analytics and Now-casting: A Comprehensive Model for Eventuality of Forecasting and Predictive Policies of Policy-making Institutions
This research aims to provide a comprehensive model of now-casting and eventuality of predictive policies based on big data analytics for policy-making institutions and indicates that the dimensions of the comprehensive model include: the alignment ofnow-casting strategies and the big data Analytics’ architecture.
Research Challenges in Financial Data Modeling and Analysis
Significant research challenges must be addressed in the cleaning, transformation, integration, modeling, and analytics of Big Data sources for finance and are of interest to data-driven financial institutions in both corporate finance and consumer finance.


Climate Predictions with Multimodel Ensembles
Abstract Several methods of combining individual forecasts from a group of climate models to produce an ensemble forecast are considered. These methods are applied to an ensemble of 500-hPa
Tracking climate models
This work uses an algorithm due to Monteleoni and Jaakkola that models the sequence of observations using a hierarchical learner, based on a set of generalized Hidden Markov Models, where the identity of the current best climate model is the hidden variable.
An analysis of multimodel ensemble predictions for seasonal climate anomalies
[1] In this paper the potential advantages and relative performances of different techniques for constructing multimodel ensemble seasonal predictions are examined. Two commonly used methods of
The rationale behind the success of multi-model ensembles in seasonal forecasting — I. Basic concept
It is demonstrated that multi-model superiority is caused not only by error compensation but in particular by its greater consistency and reliability.
A Bayesian Framework for Multimodel Regression
A framework based on Bayesian regression and constrained least squares methods for incorporating prior beliefs in a linear regression problem and new generalizations of ridge regression based on priors relevant to multimodel combinations are presented.
Risks of Model Weighting in Multimodel Climate Projections
The results indicate that for many applications equal weighting may be the safer and more transparent way to combine models, and also within the presented framework eliminating models from an ensemble can be justified if they are known to lack key mechanisms that are indispensable for meaningful climate projections.
Predictions of Nino3.4 SST in CFSv1 and CFSv2: a diagnostic comparison
Diagnostic evaluations of the relative performances of CFSv1 and CFSv2 in prediction of monthly anomalies of the ENSO-related Nino3.4 SST index are conducted using the common hindcast period of
Online Learning of Non-stationary Sequences
This work derives upper and lower relative loss bounds for a class of universal learning algorithms involving a switching dynamics over the choice of the experts and provides the optimal a priori discretization for learning the parameter that governs the switching dynamics.
Online learning of nonstation - ary sequences , ” in
  • NIPS ’ 03 : Advances in Neural Information Processing Systems
  • 2003
and W
  • Wang, “An improved in situ and satellite SST analysis for climate,” J. Climate, vol. 15, pp. 1609–1625
  • 2002