Modeling and forecasting mortality rates

Daniel Mitchell, Patrick L Brockett, Rafael Mendoza-Arriaga, Kumar Muthuraman

    Research output: Contribution to journalArticle

    35 Citations (Scopus)

    Abstract

    We show that by modeling the time series of mortality rate changes rather than mortality rate levels we can better model human mortality. Leveraging on this, we propose a model that expresses log mortality rate changes as an age group dependent linear transformation of a mortality index. The mortality index is modeled as a Normal Inverse Gaussian. We demonstrate, with an exhaustive set of experiments and data sets spanning 11 countries over 100 years, that the proposed model significantly outperforms existing models. We further investigate the ability of multiple principal components, rather than just the first component, to capture differentiating features of different age groups and find that a two component NIG model for log mortality change best fits existing mortality rate data.

    Original languageEnglish (US)
    Pages (from-to)275-285
    Number of pages11
    JournalInsurance: Mathematics and Economics
    Volume52
    Issue number2
    DOIs
    StatePublished - Mar 1 2013

    Fingerprint

    Mortality Rate
    Mortality
    Forecasting
    Modeling
    Inverse Gaussian
    Component Model
    Linear transformation
    Principal Components
    Model
    Express
    Time series
    Mortality rate
    Dependent
    Demonstrate
    Experiment
    Age groups

    Keywords

    • Mortality forecasting
    • Mortality rates
    • Statistics
    • Time series

    ASJC Scopus subject areas

    • Statistics and Probability
    • Economics and Econometrics
    • Statistics, Probability and Uncertainty

    Cite this

    Mitchell, D., Brockett, P. L., Mendoza-Arriaga, R., & Muthuraman, K. (2013). Modeling and forecasting mortality rates. Insurance: Mathematics and Economics, 52(2), 275-285. https://doi.org/10.1016/j.insmatheco.2013.01.002

    Modeling and forecasting mortality rates. / Mitchell, Daniel; Brockett, Patrick L; Mendoza-Arriaga, Rafael; Muthuraman, Kumar.

    In: Insurance: Mathematics and Economics, Vol. 52, No. 2, 01.03.2013, p. 275-285.

    Research output: Contribution to journalArticle

    Mitchell, D, Brockett, PL, Mendoza-Arriaga, R & Muthuraman, K 2013, 'Modeling and forecasting mortality rates', Insurance: Mathematics and Economics, vol. 52, no. 2, pp. 275-285. https://doi.org/10.1016/j.insmatheco.2013.01.002
    Mitchell D, Brockett PL, Mendoza-Arriaga R, Muthuraman K. Modeling and forecasting mortality rates. Insurance: Mathematics and Economics. 2013 Mar 1;52(2):275-285. https://doi.org/10.1016/j.insmatheco.2013.01.002
    Mitchell, Daniel ; Brockett, Patrick L ; Mendoza-Arriaga, Rafael ; Muthuraman, Kumar. / Modeling and forecasting mortality rates. In: Insurance: Mathematics and Economics. 2013 ; Vol. 52, No. 2. pp. 275-285.
    @article{a04b8087099d45e5af7d1451181ccf53,
    title = "Modeling and forecasting mortality rates",
    abstract = "We show that by modeling the time series of mortality rate changes rather than mortality rate levels we can better model human mortality. Leveraging on this, we propose a model that expresses log mortality rate changes as an age group dependent linear transformation of a mortality index. The mortality index is modeled as a Normal Inverse Gaussian. We demonstrate, with an exhaustive set of experiments and data sets spanning 11 countries over 100 years, that the proposed model significantly outperforms existing models. We further investigate the ability of multiple principal components, rather than just the first component, to capture differentiating features of different age groups and find that a two component NIG model for log mortality change best fits existing mortality rate data.",
    keywords = "Mortality forecasting, Mortality rates, Statistics, Time series",
    author = "Daniel Mitchell and Brockett, {Patrick L} and Rafael Mendoza-Arriaga and Kumar Muthuraman",
    year = "2013",
    month = "3",
    day = "1",
    doi = "10.1016/j.insmatheco.2013.01.002",
    language = "English (US)",
    volume = "52",
    pages = "275--285",
    journal = "Insurance: Mathematics and Economics",
    issn = "0167-6687",
    publisher = "Elsevier",
    number = "2",

    }

    TY - JOUR

    T1 - Modeling and forecasting mortality rates

    AU - Mitchell, Daniel

    AU - Brockett, Patrick L

    AU - Mendoza-Arriaga, Rafael

    AU - Muthuraman, Kumar

    PY - 2013/3/1

    Y1 - 2013/3/1

    N2 - We show that by modeling the time series of mortality rate changes rather than mortality rate levels we can better model human mortality. Leveraging on this, we propose a model that expresses log mortality rate changes as an age group dependent linear transformation of a mortality index. The mortality index is modeled as a Normal Inverse Gaussian. We demonstrate, with an exhaustive set of experiments and data sets spanning 11 countries over 100 years, that the proposed model significantly outperforms existing models. We further investigate the ability of multiple principal components, rather than just the first component, to capture differentiating features of different age groups and find that a two component NIG model for log mortality change best fits existing mortality rate data.

    AB - We show that by modeling the time series of mortality rate changes rather than mortality rate levels we can better model human mortality. Leveraging on this, we propose a model that expresses log mortality rate changes as an age group dependent linear transformation of a mortality index. The mortality index is modeled as a Normal Inverse Gaussian. We demonstrate, with an exhaustive set of experiments and data sets spanning 11 countries over 100 years, that the proposed model significantly outperforms existing models. We further investigate the ability of multiple principal components, rather than just the first component, to capture differentiating features of different age groups and find that a two component NIG model for log mortality change best fits existing mortality rate data.

    KW - Mortality forecasting

    KW - Mortality rates

    KW - Statistics

    KW - Time series

    UR - http://www.scopus.com/inward/record.url?scp=84873276598&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84873276598&partnerID=8YFLogxK

    U2 - 10.1016/j.insmatheco.2013.01.002

    DO - 10.1016/j.insmatheco.2013.01.002

    M3 - Article

    VL - 52

    SP - 275

    EP - 285

    JO - Insurance: Mathematics and Economics

    JF - Insurance: Mathematics and Economics

    SN - 0167-6687

    IS - 2

    ER -