SEARCH

SEARCH BY CITATION

Keywords:

  • times series;
  • forecasting;
  • penalized spline regression;
  • updating;
  • Gaussian process regression;
  • prediction error;
  • nonuniformly spaced data

Abstract

A statistical learning methodology, called flow field forecasting, is presented for predicting the future of a univariate time series. Flow field forecasting draws information from an interpolated flow field of the observed time series to incrementally build a forecast. The time series need not have uniformly spaced observations. Included in the presentation are measures of assessment, a procedure for forecast updating as new data arrive and a performance comparison of flow field forecasting with other major forecasting techniques. © 2013 Wiley Periodicals, Inc. Statistical Analysis and Data Mining, 2013