Fast Estimation of the Composite Link Model for Multidimensional Grouped Counts

TL;DR

This paper introduces a fast, efficient iterative estimation method for the Composite Link Model tailored for high-dimensional, grouped count data, significantly improving computational performance and practical usability.

Contribution

It reformulates the estimation process using Generalized Linear Array Models, enabling scalable and smooth estimation of latent distributions in multidimensional data.

Findings

Demonstrates improved computational speed over traditional algorithms.

Shows effective disaggregation and pattern capture in high-dimensional mortality data.

Achieves better storage efficiency and practical applicability.

Abstract

This paper presents a significant advancement in the estimation of the Composite Link Model within a penalized likelihood framework, specifically designed to address indirect observations of grouped count data. While the model is effective in these contexts, its application becomes computationally challenging in large, high-dimensional settings. To overcome this, we propose a reformulated iterative estimation procedure that leverages Generalized Linear Array Models, enabling the disaggregation and smooth estimation of latent distributions in multidimensional data. Through simulation studies and applications to high-dimensional mortality datasets, we demonstrate the model's capability to capture fine-grained patterns while comparing its computational performance to the conventional algorithm. The proposed methodology offers notable improvements in computational speed, storage efficiency, and practical applicability, making it suitable for a wide range of fields in which high-dimensional data are provided in grouped formats.