PreprintCC BY 4.0HAL
Inference for Generalized Almost Cyclostationary Processes with Unknown Lag-Dependent Cycle Frequencies
Abstract
In this paper, we study the estimation problem of generalized cyclic autocorrelation functions with unknown lag-dependent cycle frequencies for generalized almost cyclostationary processes. We establish the conditions under which the estimator is mean-square consistent and asymptotically normal. In addition, we consider the discrete-time counterpart of this estimator for sampled stochastic processes and establish the sufficient conditions for its mean-square consistency and asymptotic normality. A Monte Carlo study in a realistic simulation scenario illustrates the results: the correlogram with estimated frequencies behaves like the correlogram with the true frequencies.
Related
- Inference for sparsely sampled Gauss-Markov processes under outcome-dependent dropout
- Generalized Wishart processes under nonlinear expectation
- Random convergence of generalized Steiner processes
- Real-time and adaptive anomaly detection algorithm for cyclostationary models
- Novel Unified Variants, Properties, and Applications of Ostrowski, Jensen, and Hermite–Hadamard Inequalities for Generalized (η, p, h)-Convex Stochastic Processes
BibTeX
@misc{dudek2026inference,
title = {{Inference for Generalized Almost Cyclostationary Processes with Unknown Lag-Dependent Cycle Frequencies}},
author = {Anna E. Dudek and Antonio Napolitano and Jakub Rutkowski and Jakub Wojdyła},
howpublished = {HAL},
year = {2026},
month = aug,
url = {https://amu.hal.science/hal-05720542v2},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Inference for Generalized Almost Cyclostationary Processes with Unknown Lag-Dependent Cycle Frequencies
AU - Anna E. Dudek
AU - Antonio Napolitano
AU - Jakub Rutkowski
AU - Jakub Wojdyła
PY - 2026
DA - 2026/08/18
PB - HAL
UR - https://amu.hal.science/hal-05720542v2
AB - In this paper, we study the estimation problem of generalized cyclic autocorrelation functions with unknown lag-dependent cycle frequencies for generalized almost cyclostationary processes. We establish the conditions under which the estimator is mean-square consistent and asymptotically normal. In addition, we consider the discrete-time counterpart of this estimator for sampled stochastic processes and establish the sufficient conditions for its mean-square consistency and asymptotic normality. A Monte Carlo study in a realistic simulation scenario illustrates the results: the correlogram with estimated frequencies behaves like the correlogram with the true frequencies.
ER -