Grouped multivariate functional time series method: An application to mortality forecasting

Han Lin Shang; Yang Yang

doi:10.1007/978-3-319-55846-2_31

Grouped multivariate functional time series method: An application to mortality forecasting

Han Lin Shang^*, Yang Yang

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference proceeding contribution › peer-review

Abstract

Age-specific mortality rates are often disaggregated by different attributes, such as sex and state. Forecasting age-specific mortality rates at the sub-national levels may not add up to the forecasts at the national level. Further, the independent forecasts may not utilize correlation among sub-populations to improve forecast accuracy. Using Japanese mortality data, we extend the grouped univariate functional time series methods to grouped multivariate functional time series forecasting methods.

Original language	English
Title of host publication	Functional statistics and related fields
Editors	Germán Aneiros, Enea G. Bongiorno, Ricardo Cao, Philippe Vieu
Place of Publication	Cham
Publisher	Springer
Chapter	31
Pages	233-241
Number of pages	9
ISBN (Electronic)	9783319558462
ISBN (Print)	9783319558455
DOIs	https://doi.org/10.1007/978-3-319-55846-2_31
Publication status	Published - 2017
Externally published	Yes
Event	International Workshop on Functional and Operatorial Statistics (4th : 2017) - La Coruña, Spain Duration: 15 Jun 2017 → 17 Jun 2017

Publication series

Name	Contributions to statistics
Publisher	Springer

Conference

Conference	International Workshop on Functional and Operatorial Statistics (4th : 2017)
Country/Territory	Spain
City	La Coruña
Period	15/06/17 → 17/06/17

Keywords

Forecast Horizon
Aggregation Constraint
Point Forecast
Functional Principal Component
Holdout Sample

Access to Document

10.1007/978-3-319-55846-2_31

Cite this

@inproceedings{fe82369f85c8443d9b27098260c0912b,

title = "Grouped multivariate functional time series method: An application to mortality forecasting",

abstract = "Age-specific mortality rates are often disaggregated by different attributes, such as sex and state. Forecasting age-specific mortality rates at the sub-national levels may not add up to the forecasts at the national level. Further, the independent forecasts may not utilize correlation among sub-populations to improve forecast accuracy. Using Japanese mortality data, we extend the grouped univariate functional time series methods to grouped multivariate functional time series forecasting methods.",

keywords = "Forecast Horizon, Aggregation Constraint, Point Forecast, Functional Principal Component, Holdout Sample",

author = "Shang, {Han Lin} and Yang Yang",

year = "2017",

doi = "10.1007/978-3-319-55846-2_31",

language = "English",

isbn = "9783319558455",

series = "Contributions to statistics",

publisher = "Springer",

pages = "233--241",

editor = "Germ{\'a}n Aneiros and Bongiorno, {Enea G.} and Ricardo Cao and Philippe Vieu",

booktitle = "Functional statistics and related fields",

note = "International Workshop on Functional and Operatorial Statistics (4th : 2017) ; Conference date: 15-06-2017 Through 17-06-2017",

}

Shang, HL & Yang, Y 2017, Grouped multivariate functional time series method: An application to mortality forecasting. in G Aneiros, EG Bongiorno, R Cao & P Vieu (eds), Functional statistics and related fields. Contributions to statistics, Springer, Cham, pp. 233-241, International Workshop on Functional and Operatorial Statistics (4th : 2017), La Coruña, Spain, 15/06/17. https://doi.org/10.1007/978-3-319-55846-2_31

Grouped multivariate functional time series method: An application to mortality forecasting. / Shang, Han Lin; Yang, Yang.
Functional statistics and related fields. ed. / Germán Aneiros; Enea G. Bongiorno; Ricardo Cao; Philippe Vieu. Cham: Springer, 2017. p. 233-241 (Contributions to statistics).

Research output: Chapter in Book/Report/Conference proceeding › Conference proceeding contribution › peer-review

TY - GEN

T1 - Grouped multivariate functional time series method

T2 - International Workshop on Functional and Operatorial Statistics (4th : 2017)

AU - Shang, Han Lin

AU - Yang, Yang

PY - 2017

Y1 - 2017

N2 - Age-specific mortality rates are often disaggregated by different attributes, such as sex and state. Forecasting age-specific mortality rates at the sub-national levels may not add up to the forecasts at the national level. Further, the independent forecasts may not utilize correlation among sub-populations to improve forecast accuracy. Using Japanese mortality data, we extend the grouped univariate functional time series methods to grouped multivariate functional time series forecasting methods.

AB - Age-specific mortality rates are often disaggregated by different attributes, such as sex and state. Forecasting age-specific mortality rates at the sub-national levels may not add up to the forecasts at the national level. Further, the independent forecasts may not utilize correlation among sub-populations to improve forecast accuracy. Using Japanese mortality data, we extend the grouped univariate functional time series methods to grouped multivariate functional time series forecasting methods.

KW - Forecast Horizon

KW - Aggregation Constraint

KW - Point Forecast

KW - Functional Principal Component

KW - Holdout Sample

U2 - 10.1007/978-3-319-55846-2_31

DO - 10.1007/978-3-319-55846-2_31

M3 - Conference proceeding contribution

SN - 9783319558455

T3 - Contributions to statistics

SP - 233

EP - 241

BT - Functional statistics and related fields

A2 - Aneiros, Germán

A2 - Bongiorno, Enea G.

A2 - Cao, Ricardo

A2 - Vieu, Philippe

PB - Springer

CY - Cham

Y2 - 15 June 2017 through 17 June 2017

ER -

Grouped multivariate functional time series method: An application to mortality forecasting

Abstract

Publication series

Conference

Keywords

Access to Document

Fingerprint

Cite this