Bootstrapping Nonlinear Storm Event Modelsby Bryson C. Bates, CSIRO,
S. Ganeshanandam, CSIRO,
Abstract: A bootstrap method for estimating the parameters of nonlinear storm event models is presented. The method pools (combines) the parameter estimates obtained from fitting computed and observed flood hydrographs for several storm events on the watershed. The pooled mean makes explicit allowance for parameter estimates of varying precision. Unlike parametric pooling techniques, the method does not make any distributional assumptions about the model parameters. An example involving a two -- parameter, nonlinear surface runoff model demonstrates the potential advantages of bootstrapping over parametric pooling techniques.
Subject Headings: Parameters (statistics) | Storms | Nonlinear analysis | Watersheds | Runoff | Floods | Hydrographs | Terrain models
Services: Buy this book/Buy this article
Return to search