Dresden 2014 – scientific programme

Parts | Days | Selection | Search | Updates | Downloads | Help

DY: Fachverband Dynamik und Statistische Physik

DY 17: Modeling and Data Analysis

DY 17.7: Talk

Wednesday, April 2, 2014, 11:15–11:30, ZEU 146

Optimal model-free prediction of multivariate time series — •Jakob Runge — Potsdam Institute for Climate Impact Research, Potsdam, Germany, and Humboldt University Berlin

We address the problem of predicting a single time series from a set of multivariate predictors in an information theoretic framework. We investigate in how far this can be done optimally given the available information and develop a practical prediction algorithm. The performance and challenges are demonstrated on multivariate nonlinear stochastic delay processes as well as on real data.