Clustering and forecasting multiple functional time series

DOI10.1214/22-AOAS1602zbMATH Open1498.62327arXiv2201.01024OpenAlexW4226514250MaRDI QIDQ2080765FDOQ2080765

Authors: Chen Tang, Han Lin Shang, Yanrong Yang 

Publication date: 10 October 2022

Published in: The Annals of Applied Statistics (Search for Journal in Brave)

Abstract: Modelling and forecasting homogeneous age-specific mortality rates of multiple countries could lead to improvements in long-term forecasting. Data fed into joint models are often grouped according to nominal attributes, such as geographic regions, ethnic groups, and socioeconomic status, which may still contain heterogeneity and deteriorate the forecast results. Our paper proposes a novel clustering technique to pursue homogeneity among multiple functional time series based on functional panel data modelling to address this issue. Using a functional panel data model with fixed effects, we can extract common functional time series features. These common features could be decomposed into two components: the functional time trend and the mode of variations of functions (functional pattern). The functional time trend reflects the dynamics across time, while the functional pattern captures the fluctuations within curves. The proposed clustering method searches for homogeneous age-specific mortality rates of multiple countries by accounting for both the modes of variations and the temporal dynamics among curves. We demonstrate that the proposed clustering technique outperforms other existing methods through a Monte Carlo simulation and could handle complicated cases with slow decaying eigenvalues. In empirical data analysis, we find that the clustering results of age-specific mortality rates can be explained by the combination of geographic region, ethnic groups, and socioeconomic status. We further show that our model produces more accurate forecasts than several benchmark methods in forecasting age-specific mortality rates.

Full work available at URL: https://arxiv.org/abs/2201.01024

Recommendations

• Common functional principal component models for mortality forecasting
• Forecasting multiple functional time series in a group structure: an application to mortality
• Robust forecasting of mortality and fertility rates: a functional data approach
• Incorporating statistical clustering methods into mortality models to improve forecasting performances
• Coherent mortality forecasting by the weighted multilevel functional principal component approach

zbMATH Keywords

functional principal component analysisfunctional time seriesage-specific mortality forecastingfunctional panel datamultilevel functional data

Mathematics Subject Classification ID

Factor analysis and principal components; correspondence analysis (62H25) Classification and discrimination; cluster analysis (statistical aspects) (62H30) Time series, auto-correlation, regression, etc. in statistics (GARCH) (62M10) Inference from stochastic processes and prediction (62M20) Applications of statistics to social sciences (62P25) Functional data analysis (62R10)

Cites Work

• Model-based clustering of time series in group-specific functional subspaces
• The discriminative functional mixture model for a comparative analysis of bike sharing systems
• Large Sample Properties of Generalized Method of Moments Estimators
• Dynamic functional principal components
• Multilevel functional principal component analysis
• Modeling and forecasting U.S. mortality. (With discussion)
• Clustering functional data
• Functional data clustering: a survey
• Are more data always better for factor analysis?
• Inference for functional data with applications
• Functional Clustering and Identifying Substructures of Longitudinal Data
• Strictly Proper Scoring Rules, Prediction, and Estimation
• Functional Data Analysis for Sparse Longitudinal Data
• Title not available (Why is that?)
• Finding the Number of Clusters in a Dataset
• Robust forecasting of mortality and fertility rates: a functional data approach
• High-dimensional data clustering
• Heteroskedasticity and Autocorrelation Consistent Covariance Matrix Estimation
• Forecasting functional time series
• Econometric analysis of cross section and panel data.
• Fourier analysis of stationary time series in function space
• Functional Principal Component Regression and Functional Partial Least Squares
• A Simple, Positive Semi-Definite, Heteroskedasticity and Autocorrelation Consistent Covariance Matrix
• Title not available (Why is that?)
• On the Concept of Depth for Functional Data
• Panel data models with interactive fixed effects
• Title not available (Why is that?)
• A proposal for robust curve clustering
• Analysis of panel data
• On the prediction of stationary functional time series
• Model-based clustering for multivariate functional data
• Dynamical functional prediction and classification, with application to traffic flow prediction
• Generalized multilevel functional regression
• Unsupervised Curve Clustering using B‐Splines
• Functional Modelling and Classification of Longitudinal Data*
• Title not available (Why is that?)
• Estimation of the Mean of Functional Time Series and a Two-Sample Problem
• Assessing the Finite Dimensionality of Functional Data
• Selecting the number of principal components in functional data
• Smoothing and forecasting mortality rates
• Lee-Carter mortality forecasting with age-specific enhancement.
• CATS
• Multivariate density estimation with general flat-top kernels of infinite order
• Title not available (Why is that?)
• A plug-in bandwidth selection procedure for long-run covariance estimation with stationary functional time series
• On flat-top kernel spectral density estimators for homogeneous random fields
• Bootstrap methods for stationary functional time series
• Mortality and life expectancy forecasting for a group of populations in developed countries: a multilevel functional data method
• A note on estimation in Hilbertian linear models
• Multilevel functional clustering analysis
• Linear statistical models.
• Longitudinal functional principal component analysis

Cited In (6)

• Forecasting Simultaneously High‐Dimensional Time Series: A Robust Model‐Based Clustering Approach
• Time series clustering based on forecast densities
• Incorporating statistical clustering methods into mortality models to improve forecasting performances
• Detection and estimation of structural breaks in high-dimensional functional time series
• Non-linear time series clustering based on non-parametric forecast densities
• Multivariate functional data modeling with time-varying clustering

Uses Software

• Rainbow
• WinBUGS
• forecast
• Demography
• funHDDC
• Human Mortality
• ftsa
• MTS

This page was built for publication: Clustering and forecasting multiple functional time series

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2080765)