Towards unbiased reconstruction of noisy chaotic dynamics using a robust hypersurface minimization technique

The method of least squares (LS) and total least,squares (TLS) are two methods widely used in the application of best-fit curve, but they generally provide biased results especially when the structure is nonlinear. To overcome the inherent limitations of both LS and TLS methods, we present a new method that is based on minimizing hypersurface distance. Computer simulation examples show that the new method proposed achieves more accurate parameter estimates than either the LS and TLS.