Academic Journal

Robust scale estimation for strongly mixing processes under shifts in the mean.

Λεπτομέρειες βιβλιογραφικής εγγραφής
Τίτλος: Robust scale estimation for strongly mixing processes under shifts in the mean.
Συγγραφείς: Axt, Ieva1 (AUTHOR), Fried, Roland1 (AUTHOR) fried@statistik.tu-dortmund.de
Πηγή: Statistics. Aug2026, Vol. 60 Issue 4, p1240-1258. 19p.
Θεματικοί όροι: *Standard deviations, *Outliers (Statistics), *Change-point problems, *Simulation methods & models, *Ergodic theory, *Estimation theory
Περίληψη: Knowledge of the variance or standard deviation of a process is crucial in many applications, e.g., for standardization or assessment of uncertainty. Traditional estimators of scale, like the sample variance, exhibit significant bias when faced with outliers or shifts in the mean. In this work, we establish the strong consistency of a blockwise modification of the MAD for strongly mixing processes. A simulation study is conducted to examine the properties of this estimation approach in the possible presence of outliers and level shifts. [ABSTRACT FROM AUTHOR]
Βάση Δεδομένων: Academic Search Index
Περιγραφή
ISSN:02331888
DOI:10.1080/02331888.2025.2600463