×

Outlier analysis and mortality forecasting: the United Kingdom and Scandinavian countries. (English) Zbl 1092.91050

The authors perform a systematic time-series outlier analysis of the time-varying parameter, often known as the mortality index, encompassed in the Lee-Carter mortality forecasting model. The analysis begins with identifying exogenous events that might significantly affect the mortality dynamics of selected countries. The authors do this using Chen and Liu’s method to search iteratively for outliers present on the mortality index, and then matching the identified outliers with events that may have caused them. They re-estimate the model parameters during the outlier detection process and introduce an additional iterative cycle aiming at unveiling the true model underling the outlier-free series of the mortality index. It is scrutinized the efficiency of the outlier-adjusted model by comparing the interval forecasts of the mortality index, and related variables such as the central rates of death and the complete life expectancies at birth. Several potential limitations of the Lee-Carter model in forecasting mortality are presented.

MSC:

91B30 Risk theory, insurance (MSC2010)
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] Avian Influenza – Fact Sheet (2004)
[2] Rowley MC, Record, Society of Actuaries 28 (2001)
[3] Wolak DL, Record. Society of Actuaries 29 pp 1– (2003)
[4] DOI: 10.2307/2287980 · doi:10.2307/2287980
[5] DOI: 10.2307/2290201 · doi:10.2307/2290201
[6] DOI: 10.1016/0169-2070(92)90055-E · doi:10.1016/0169-2070(92)90055-E
[7] DOI: 10.1017/S1357321700000155 · doi:10.1017/S1357321700000155
[8] Chan WS, Journal of Actuarial Practice 6 pp 149– (1998)
[9] DOI: 10.1002/jae.3950090205 · doi:10.1002/jae.3950090205
[10] DOI: 10.1016/S0167-9473(01)00014-7 · Zbl 1079.62542 · doi:10.1016/S0167-9473(01)00014-7
[11] DOI: 10.1002/sim.4780111610 · doi:10.1002/sim.4780111610
[12] DOI: 10.1016/S0167-6687(03)00118-5 · Zbl 1025.62041 · doi:10.1016/S0167-6687(03)00118-5
[13] DOI: 10.1111/1467-9876.00393 · Zbl 1111.62359 · doi:10.1111/1467-9876.00393
[14] DOI: 10.1017/S1474747203001239 · doi:10.1017/S1474747203001239
[15] Booth H Tickle L 2003 The future aged: new projections of Australia’s elderly population Department of Actuarial Studies Research Paper Series. Macquarie University, Sydney, Australia
[16] Buettner T, North American Actuarial Journal 6 pp 14– (2002) · Zbl 1084.62524 · doi:10.1080/10920277.2002.10596053
[17] Tabeau E, Forecasting Mortality in Developed Countries – Insights from a Statistical, Demographic and Epidemiological Perspective (2001)
[18] Wong-Fupuy C, North American Actuarial Journal 8 pp 56– (2004) · Zbl 1085.62517 · doi:10.1080/10920277.2004.10596137
[19] DOI: 10.1080/034612301750077374 · Zbl 0973.62106 · doi:10.1080/034612301750077374
[20] Fledelius P, Journal of Actuarial Practice 11 pp 103– (2004)
[21] DOI: 10.1080/034612301101016516 · Zbl 1142.62078 · doi:10.1080/034612301101016516
[22] DOI: 10.2307/2290724 · Zbl 0775.62229 · doi:10.2307/2290724
[23] DOI: 10.1002/for.3980120103 · doi:10.1002/for.3980120103
[24] DOI: 10.1016/0378-4754(94)00094-7 · doi:10.1016/0378-4754(94)00094-7
[25] DOI: 10.1038/35015561 · doi:10.1038/35015561
[26] Human Mortality Database University of California, Berkeley (USA), and Max Planck Institute for Demographic Research (Germany). Available at,www.mortality.orgwww.humanmortality.de, (data downloaded on 25 Jan 2004)
[27] Mortality statistics: general – DH1 no. 33 (1999)
[28] Mortality statistics: general – DH1 no. 33 (2000)
[29] Social Trends. no. 31 (2001)
[30] Glei D, About mortality data for Sweden (2003)
[31] DOI: 10.2307/3644567 · doi:10.2307/3644567
[32] Box GEP, Time series analysis forecasting and control (1976)
[33] Payne RW, Genstat 5 Release 3 Reference Manual (1993)
[34] Brose GJ, Numerical Methods with MATLAB: A Resource for Scientists and Engineers (1997)
[35] IMSL Math/Library: Fortran Subroutines for Mathematical Applications. Volume 1. (1997)
[36] DOI: 10.1016/S0167-6687(02)00185-3 · Zbl 1074.62524 · doi:10.1016/S0167-6687(02)00185-3
[37] Wilmoth JR 1993 Computational methods for fitting and extrapolating the Lee-Carter model of mortality change Technical report. Department of Demography. University of California, Berkeley
[38] Lee RD Nault F 1993 Modeling and forecasting provincial mortality in Canada Presented at the World Congress of the International Union for the Scientific Study of Population, August 24-September 1, Montreal
[39] DOI: 10.1353/dem.2001.0036 · doi:10.1353/dem.2001.0036
[40] Booth H Maindonald J Smith L 2002 Age-time interactions in mortality projection: applying Lee-Carter to Australia Working papers in demography. Demography and Sociology Program, Research School of Social Sciences. Australian National University
[41] Carter LR Prskawetz A 2001 Examining structural shifts in mortality using the Lee-Carter method MPIDR working paper WP 2001–007. Max Planck Institute for Demographic Research
[42] Charlton J, The Health of Adult Britain 1841–1994 1 (1997)
[43] Fox AJ, Journal of the Royal Statistical Society Series B 34 pp 350– (1972)
[44] DOI: 10.2307/1270165 · doi:10.2307/1270165
[45] DOI: 10.2307/1268821 · Zbl 0693.62072 · doi:10.2307/1268821
[46] Muirhead CR, Journal of the Royal Statistical Society Series B 48 pp 39– (1986)
[47] Ljung GM, Journal of the Royal Statistical Society Series B 55 pp 559– (1993)
[48] DOI: 10.1002/for.3980070102 · doi:10.1002/for.3980070102
[49] DOI: 10.1002/(SICI)1099-131X(200001)19:1<23::AID-FOR740>3.0.CO;2-2 · doi:10.1002/(SICI)1099-131X(200001)19:1<23::AID-FOR740>3.0.CO;2-2
[50] Wei WWS, Time series analysis: univariate and multivariate methods (1990)
[51] Liu LM, Forecasting and Time Series Analysis Using the SCA Statistical System (1994)
[52] DOI: 10.2307/1391755 · doi:10.2307/1391755
[53] DOI: 10.2307/2685050 · Zbl 04533950 · doi:10.2307/2685050
[54] What’s new in SAS/ETS 9? (2004)
[55] Charlton J, Monitoring Health – data sources and methods 1 (1997)
[56] Table 4: Armed Forces Mobilized and Casualties in World War I (2004)
[57] World War I (2004)
[58] DOI: 10.1126/science.1096550 · doi:10.1126/science.1096550
[59] Keilman N, Journal of Official Statistics (Statistics Sweden) 13 pp 245– (1998)
[60] DOI: 10.1093/biomet/65.2.297 · Zbl 0386.62079 · doi:10.1093/biomet/65.2.297
[61] DOI: 10.1109/TAC.1974.1100705 · Zbl 0314.62039 · doi:10.1109/TAC.1974.1100705
[62] DOI: 10.2307/1391361 · doi:10.2307/1391361
[63] DOI: 10.1016/0169-2070(89)90090-3 · doi:10.1016/0169-2070(89)90090-3
[64] DOI: 10.1002/for.3980030208 · doi:10.1002/for.3980030208
[65] DOI: 10.1017/S1357321700003822 · doi:10.1017/S1357321700003822
[66] Sarony N, Personal Injury Tables Hong Kong: Tables for the Calculation of Damages (2003)
[67] DOI: 10.1111/1467-985X.00119 · Zbl 04549845 · doi:10.1111/1467-985X.00119
[68] Li SH, Journal of Actuarial Practice 11 pp 129– (2004)
[69] Tong H, Threshold Models in Non-Linear Time Series Analysis (1983) · Zbl 0527.62083 · doi:10.1007/978-1-4684-7888-4
[70] Willekens FJ, Emerging Issues in Demographic Research pp pp. 9–75– (1990)
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.