Statistical Inference on Hierarchical Simultaneous Autoregressive Models with Missing Data

• DOI: 10.48550/arxiv.2403.17257
• Publication Date: 2024-03-25
• Authors: Anjana Wijayaward...; Thomas Suesse; David Gunawan
• Keywords: Statistical Inference

Efficient estimation methods for simultaneous autoregressive (SAR) models with missing data in the response variable have been well-developed literature. It is common practice to introduce a measurement error into SAR models. The serves distinguish noise component from spatial process. However, previous literature has not considered adding data. maximum likelihood such large datasets challenging and computationally expensive. This paper proposes two efficient likelihood-based methods: marginal (ML) expectation-maximisation (EM) algorithms estimating both errors variable. model (SEM) (SAM), popular types, are considered. mechanism assumed follow at random (MAR). While naive calculation approaches lead computational complexities of $O(n^3)$, where n total number observations, our ML EM designed reduce complexity. performance proposed investigated empirically using simulated real datasets.