EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

ROBUST LOGISTIC MAP PARAMETER ESTIMATION VIA ARCSINE-QUANTILE MATCHING

Kenneth P. Perez

Far East Journal of Mathematical Sciences (FJMS) · 22.05.2026 · 143 (6) · 1827-1838

DOI: 10.17654/0972087126106 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Far East Journal of Mathematical Sciences (FJMS), online erschienen am 22.05.2026. Verfasst von Kenneth P. Perez. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The logistic map is a canonical discrete-time model of deterministic chaos in which a single control parameter $\theta$ governs the full spectrum of dynamical behavior, from fixed-point convergence to fully developed chaos. Accurate estimation of $\theta$ from observed orbit data is essential in ecology, nonlinear electronics and secure communications. The classical estimator $\hat{\theta}_M = 4 \text{max}_i x_i$ performs well on clean data but is fundamentally fragile under noise: it relies on a single extreme order statistic, so even one outlier renders it arbitrarily inaccurate. To address this deficiency, we propose the $\bf{arcsine-quantile \, matching \,(AQM) \, estimator}$, which fits all $m = 50$ empirical orbit quantiles simultaneously to the known arcsine invariant measure by weighted least squares. AQM is computable in $O(n \, \text {log} \, n)$ time and possesses a provably bounded influence function, in sharp contrast to the unbounded influence function of the classical estimator. Monte Carlo experiments $(R = 6,000$ replications, $\theta = 4.0)$ show that AQM reduces RMSE by up to 94% over $\hat{\theta}_M$ under realistic Gaussian noise $(\sigma \ge 0.05$, unclipped $\text {data})$, while the classical estimator retains clear superiority on clean data $(\text {RMSE} < 10^{-4})$. A practical decision rule based on the empirical Lyapunov exponent $\hat {\lambda}$ is provided: use AQM when $\hat {\lambda} \ge 0.4$ and data are unclipped; revert to $\hat{\theta}_M $ when data are clean or clipped to $[0, 1]$.

Zitieren

APA

Perez, K. P. (2026). ROBUST LOGISTIC MAP PARAMETER ESTIMATION VIA ARCSINE-QUANTILE MATCHING. Far East Journal of Mathematical Sciences (FJMS), 143(6), 1827-1838. https://doi.org/10.17654/0972087126106

Vancouver

Perez KP. ROBUST LOGISTIC MAP PARAMETER ESTIMATION VIA ARCSINE-QUANTILE MATCHING. Far East Journal of Mathematical Sciences (FJMS). 2026;143(6):1827-1838. doi:10.17654/0972087126106

BibTeX
@article{Perez2026,
  author = {Perez, Kenneth P.},
  title = {ROBUST LOGISTIC MAP PARAMETER ESTIMATION VIA ARCSINE-QUANTILE MATCHING},
  journal = {Far East Journal of Mathematical Sciences (FJMS)},
  year = {2026},
  volume = {143},
  number = {6},
  pages = {1827--1838},
  doi = {10.17654/0972087126106}
}
Kostenlos starten →