Estimates of the local dynamics on a reconstructed attractor may be sensitive to the measured values of only a few points, even if many observations are used to fit the model. Bootstrap sampling provides a convenient way to identify such influential points. This article describes several applications of bootstrap sampling to quantify the uncertainty in estimates of dynamical information from data.
|Physical Review E - Statistical, Nonlinear, and Soft Matter Physics
|Published - Jul 1 2001
ASJC Scopus subject areas
- Statistical and Nonlinear Physics
- Statistics and Probability
- Condensed Matter Physics