ABSTRACT
Estimation Results Taking the essentiality that correlated integral ( )mC r varies with embedded dimension m and threshold r into account, we choose m from 1 to 20, and r takes 100 different values in interval
estimated by the use of OLS method for the following linear regression
where the negative of slope k is just ( )noisyK r . Then we apply Levenberg-Marquardt method and general global optimization method
to fi t ( )noisyK r according to (4). The estimation value of σ , NSR, decision coeffi cient of fi tting 2R and the root of mean square errors (RMSE) are given in Table 5. The statistics in Table 5 and Fig.2 all suffi ciently indicate that the fi tting makes great effect.
