Forecasting Australian subnational age-specific mortality rates

TL;DR

This paper introduces a grouped multivariate functional time series approach for forecasting Australian regional and remote age-specific mortality rates, emphasizing the importance of correlation and forecast reconciliation for improved accuracy.

Contribution

It presents a novel grouped multivariate functional time series method that enhances forecast accuracy by modeling correlations among sub-populations and ensuring consistent group-level forecasts.

Findings

Joint modeling improves point forecast accuracy.

The proposed method outperforms univariate approaches.

Forecast reconciliation ensures consistency across levels.

Abstract

When modeling sub-national mortality rates, it is important to incorporate any possible correlation among sub-populations to improve forecast accuracy. Moreover, forecasts at the sub-national level should aggregate consistently across the forecasts at the national level. In this study, we apply a grouped multivariate functional time series to forecast Australian regional and remote age-specific mortality rates and reconcile forecasts in a group structure using various methods. Our proposed method compares favorably to a grouped univariate functional time series forecasting method by comparing one-step-ahead to five-step-ahead point forecast accuracy. Thus, we demonstrate that joint modeling of sub-populations with similar mortality patterns can improve point forecast accuracy.