A new class of functional conditional autoregressive models

Abstract

We introduce a new class of conditional autoregressive models for spatially dependent functional data, formulated through conditional means, given neighbouring functional observations and characterized by a covariance operator and a spatial dependence parameter. Our estimation strategy consists of three components: (i) estimating the covariance operator using conditionally centred data, (ii) estimating the spatial dependence parameter by maximizing the likelihood of the projected observations, and (iii) applying a novel profile-based approach to obtain the final estimators. Under an expanding lattice framework, we establish two key theoretical results. First, we establish the consistency of the proposed covariance estimator, which is unattainable using naive methods based on marginally centred data. Second, we prove that the spatial dependence parameter estimator is superconsistent and asymptotically normal, where the latter property enables statistical inference for spatial dependence in functional data, a contribution that is novel in the existing literature. Numerical studies support the theoretical results and demonstrate the computational efficiency of our method. Finally, we illustrate its practical utility by analysing weekly PM2.5concentration trajectories in 2019 across counties in the Midwestern United States.