Flow field forecasting for univariate time series

A statistical learning methodology, called ow eld forecasting, is presented for predicting the future of a univariate time series. Flow eld forecasting draws information from an interpolated ow eld 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 ow eld forecasting with other major forecasting techniques. © 2013 Wiley Periodicals, Inc. Statistical Analysis and Data Mining, 2013