Smoothing two-dimensional Malaysian mortality data using P-splines indexed by age and year

Halim Shukri Kamaruddin, Noriszura Ismail

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Nonparametric regression implements data to derive the best coefficient of a model from a large class of flexible functions. Eilers and Marx (1996) introduced P-splines as a method of smoothing in generalized linear models, GLMs, in which the ordinary B-splines with a difference roughness penalty on coefficients is being used in a single dimensional mortality data. Modeling and forecasting mortality rate is a problem of fundamental importance in insurance company calculation in which accuracy of models and forecasts are the main concern of the industry. The original idea of P-splines is extended to two dimensional mortality data. The data indexed by age of death and year of death, in which the large set of data will be supplied by Department of Statistics Malaysia. The extension of this idea constructs the best fitted surface and provides sensible prediction of the underlying mortality rate in Malaysia mortality case.

Original languageEnglish
Title of host publicationAIP Conference Proceedings
PublisherAmerican Institute of Physics Inc.
Pages487-496
Number of pages10
Volume1602
ISBN (Print)9780735412361
DOIs
Publication statusPublished - 2014
Event3rd International Conference on Mathematical Sciences, ICMS 2013 - Kuala Lumpur
Duration: 17 Dec 201319 Dec 2013

Other

Other3rd International Conference on Mathematical Sciences, ICMS 2013
CityKuala Lumpur
Period17/12/1319/12/13

Fingerprint

mortality
splines
smoothing
Malaysia
death
forecasting
coefficients
penalties
regression analysis
roughness
industries
statistics
predictions

Keywords

  • Mortality data
  • P-splines
  • Smoothing

ASJC Scopus subject areas

  • Physics and Astronomy(all)

Cite this

Kamaruddin, H. S., & Ismail, N. (2014). Smoothing two-dimensional Malaysian mortality data using P-splines indexed by age and year. In AIP Conference Proceedings (Vol. 1602, pp. 487-496). American Institute of Physics Inc.. https://doi.org/10.1063/1.4882530

Smoothing two-dimensional Malaysian mortality data using P-splines indexed by age and year. / Kamaruddin, Halim Shukri; Ismail, Noriszura.

AIP Conference Proceedings. Vol. 1602 American Institute of Physics Inc., 2014. p. 487-496.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Kamaruddin, HS & Ismail, N 2014, Smoothing two-dimensional Malaysian mortality data using P-splines indexed by age and year. in AIP Conference Proceedings. vol. 1602, American Institute of Physics Inc., pp. 487-496, 3rd International Conference on Mathematical Sciences, ICMS 2013, Kuala Lumpur, 17/12/13. https://doi.org/10.1063/1.4882530
Kamaruddin HS, Ismail N. Smoothing two-dimensional Malaysian mortality data using P-splines indexed by age and year. In AIP Conference Proceedings. Vol. 1602. American Institute of Physics Inc. 2014. p. 487-496 https://doi.org/10.1063/1.4882530
Kamaruddin, Halim Shukri ; Ismail, Noriszura. / Smoothing two-dimensional Malaysian mortality data using P-splines indexed by age and year. AIP Conference Proceedings. Vol. 1602 American Institute of Physics Inc., 2014. pp. 487-496
@inproceedings{6807ef738dfb4e829ae130fe3ddeb3fe,
title = "Smoothing two-dimensional Malaysian mortality data using P-splines indexed by age and year",
abstract = "Nonparametric regression implements data to derive the best coefficient of a model from a large class of flexible functions. Eilers and Marx (1996) introduced P-splines as a method of smoothing in generalized linear models, GLMs, in which the ordinary B-splines with a difference roughness penalty on coefficients is being used in a single dimensional mortality data. Modeling and forecasting mortality rate is a problem of fundamental importance in insurance company calculation in which accuracy of models and forecasts are the main concern of the industry. The original idea of P-splines is extended to two dimensional mortality data. The data indexed by age of death and year of death, in which the large set of data will be supplied by Department of Statistics Malaysia. The extension of this idea constructs the best fitted surface and provides sensible prediction of the underlying mortality rate in Malaysia mortality case.",
keywords = "Mortality data, P-splines, Smoothing",
author = "Kamaruddin, {Halim Shukri} and Noriszura Ismail",
year = "2014",
doi = "10.1063/1.4882530",
language = "English",
isbn = "9780735412361",
volume = "1602",
pages = "487--496",
booktitle = "AIP Conference Proceedings",
publisher = "American Institute of Physics Inc.",

}

TY - GEN

T1 - Smoothing two-dimensional Malaysian mortality data using P-splines indexed by age and year

AU - Kamaruddin, Halim Shukri

AU - Ismail, Noriszura

PY - 2014

Y1 - 2014

N2 - Nonparametric regression implements data to derive the best coefficient of a model from a large class of flexible functions. Eilers and Marx (1996) introduced P-splines as a method of smoothing in generalized linear models, GLMs, in which the ordinary B-splines with a difference roughness penalty on coefficients is being used in a single dimensional mortality data. Modeling and forecasting mortality rate is a problem of fundamental importance in insurance company calculation in which accuracy of models and forecasts are the main concern of the industry. The original idea of P-splines is extended to two dimensional mortality data. The data indexed by age of death and year of death, in which the large set of data will be supplied by Department of Statistics Malaysia. The extension of this idea constructs the best fitted surface and provides sensible prediction of the underlying mortality rate in Malaysia mortality case.

AB - Nonparametric regression implements data to derive the best coefficient of a model from a large class of flexible functions. Eilers and Marx (1996) introduced P-splines as a method of smoothing in generalized linear models, GLMs, in which the ordinary B-splines with a difference roughness penalty on coefficients is being used in a single dimensional mortality data. Modeling and forecasting mortality rate is a problem of fundamental importance in insurance company calculation in which accuracy of models and forecasts are the main concern of the industry. The original idea of P-splines is extended to two dimensional mortality data. The data indexed by age of death and year of death, in which the large set of data will be supplied by Department of Statistics Malaysia. The extension of this idea constructs the best fitted surface and provides sensible prediction of the underlying mortality rate in Malaysia mortality case.

KW - Mortality data

KW - P-splines

KW - Smoothing

UR - http://www.scopus.com/inward/record.url?scp=84904124369&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84904124369&partnerID=8YFLogxK

U2 - 10.1063/1.4882530

DO - 10.1063/1.4882530

M3 - Conference contribution

AN - SCOPUS:84904124369

SN - 9780735412361

VL - 1602

SP - 487

EP - 496

BT - AIP Conference Proceedings

PB - American Institute of Physics Inc.

ER -