Nearest‐neighbor methods for nonparametric rainfall‐runoff forecasting
Authors:
- M. KarlssonMKM. Karlsson
- S. YakowitzSYS. Yakowitz
Open Access
- 9 July 1987
- journal article
- Published by American Geophysical Union (AGU) in Water Resources Research
- Vol. 23 (7) , 1300-1308
Abstract
This paper introduces a nonparametric regression method for time series to the stochastic hydrology literature. The technique, known as the nearest‐neighbor (NN) method, is motivated by the classical rainfall‐runoff forecasting problem. We have shown in a companion paper (Yakowitz, 1987) that in the time series setting it has attractive large‐sample properties. As summarized herein, these include asymptotic convergence to the optimal forecaster under many circumstances in which the forecaster is not a linear function of the data. The technique is applied to a sequence of about 3000 seasonalized rainfall‐runoff pairs from the Coshocton Watershed. We compare the NN forecasting performance on a split sample with those of autoregressive maximum and instantaneous unit hydrograph forecasters.Keywords
This publication has 7 references indexed in Scilit:
- NEAREST‐NEIGHBOUR METHODS FOR TIME SERIES ANALYSISJournal of Time Series Analysis, 1987
- Nearest Neighbor Methods for Time Series, with Application to Rainfall/Runoff Prediction1987
- PARAMETRIC/NONPARAMETRIC MIXTURE DENSITY ESTIMATION WITH APPLICATION TO FLOOD‐FREQUENCY ANALYSIS1Jawra Journal of the American Water Resources Association, 1985
- Nonparametric Density Estimation, Prediction, and Regression for Markov SequencesJournal of the American Statistical Association, 1985
- Markov Flow Models and the Flood Warning ProblemWater Resources Research, 1985
- Optimal Rates of Convergence for Nonparametric EstimatorsThe Annals of Statistics, 1980
- Multivariate k-nearest neighbor density estimatesJournal of Multivariate Analysis, 1979