Forecasting Australian subnational age-specific mortality rates
- PDF / 1,858,279 Bytes
- 24 Pages / 439.37 x 666.142 pts Page_size
- 41 Downloads / 196 Views
Forecasting Australian subnational age‑specific mortality rates Han Lin Shang1 · Yang Yang2 Accepted: 22 October 2020 © Springer Nature B.V. 2020
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. Keywords Multivariate functional principal component analysis · Hierarchical/ grouped time series · Forecast reconciliation · Australian regional mortality rates
Introduction Mortality is regarded as one of the most widely available measures of health conditions in a community. National and sub-national mortality patterns in terms of age, sex, and geographical distribution are of interest to epidemiologists, health care personnel, and those working in health and social policy, planning, and administration. Traditional studies of human mortality focus on population data at the national level. In recent years, academics and government stakeholders have been showing increasing interest in regional mortality improvements after realizing that sub-national forecasts of age-specific mortality rates are useful for informing social and economic * Han Lin Shang [email protected] 1
Department of Actuarial Studies and Business Analytics, Macquarie University, Level 7, 4 Eastern Rd, Sydney, NSW 2109, Australia
2
Research School of Finance, Actuarial Studies and Statistics, Australian National University, Canberra, ACT 2601, Australia
13
Vol.:(0123456789)
H. L. Shang, Y. Yang
policies within local regions. Thus, any improvement in the forecast accuracy of regional mortality rates would help determine the allocation of current and future resources at the national and sub-national levels. Many mortality modeling methods have been proposed since the publication of the Gompertz law in 1825 (see Booth and Tickle 2008, for a comprehensive literature review on mortality modeling and forecasting), but only a few approaches can forecast multiple regional mortality rates within a country simultaneously. Thus, proposing a multivariate functional time series method for producing accurate and coherent sub-national age-specific mortality forecasts is the primary motivation of our study. Functional time series can arise by separating an almost continuous time record into consecutive intervals, such as the intraday trajectories of the S&P 500 index (Shang 2017) and the monthly sea surface tempera
Data Loading...
