Parameter estimation for fractional autoregressive process with periodic structure

Abstract: This paper introduces a new kind of periodic fractional autoregressive process (PFAR) driven by fractional Gaussian noise (fGn). The new model is a specialized varying coefficient fractional autoregressive model, where the coefficients adhere to a periodic structure. In this working, Generalized least squares estimation and GPH method are employed to construct an initial estimator to estimate the joint estimation of the parameters of these models. Then one-step procedure is used to obtain a more asymptotically-efficient estimator. The paper proves that both estimators are consistent and asymptotically normal, and their performance is demonstrated through a simulation study using finite-size samples via Monte Carlo simulations. Simulation studies suggests that, while both estimation methods can accurately estimate the model, the one-step estimator outperforms the initial estimator.