Abstract
In this paper, we focus on Lee–Carter mortality forecasting. Model residuals and future mortality trendsare explored. Predictions of the force of mortality for France, Belarus and Lithuania are provided and compared. Severalmodifications of the model are applied to Lithuanian mortality data in order to obtain the most precise forecast.
Highlights
Mortality is not constant over time; it changes differently in different age groups
Let us consider that kt is a random process with two lags, i.e. a third-order autoregressive process with a drift: kt = θkt−1 + θ0 + θ1 kt−1 − kt−2 + θ2 kt−2 − kt−3 + ξt, t = 1971, ..., 2005, with independent Gaussian errors ξt ∼ N(0, σ2rw) and θ = 1 (p-values of the augmented Dickey–Fuller (ADF) test are 0.2774 for men and 0.6371 for women)
While comparing the suitability of the Lee–Carter model for different countries, we have obtained that the model most accurately describes and predicts mortality for France
Summary
Mortality is not constant over time; it changes differently in different age groups. The Lee–Carter model has been the most widely used one This model was originally applied in 1992 to US data and later to mortality data of many other countries such as Canada (Nault, 1993), Austria (Carter and Prkawetz, 2001), Italy (Giacometti, Bertocchi, Rachev, Fabozzi, 2012). Lithuanian mortality has not been predicted; the objective of this paper is to apply the Lee–Carter model to Lithuanian mortality data and identify the best fitting modification. The paper compares the applicability of the model to populations with different mortality profiles. For this purpose, besides Lithuania, France and Belarus were chosen
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.