Stochastic modelling of actuarial assumptions using Chinese data
Date
2015
Authors
Huang,Fei
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
In this thesis, we develop stochastic economic and mortality models for actuarial use in China. Firstly, we conduct the first study of stochastic economic modelling with Chinese data for actuarial use. Univariate models, vector autoregression (VAR), and two cascade systems are described and compared. We focus on six major economic assumptions for modelling purposes. Granger causality tests are used to identify the driving force of a cascade system. Robust standard errors are estimated for each model. Diagnostic checking of residuals, goodness-of-fit measures and out-of-sample validations are applied for model selection. By comparing different models for each variable, we find that the equity-driving cascade system is the best structure for actuarial use in China. The forecasts of the variables could be applied as economic inputs to stochastic projection models of insurance portfolios or pension funds for short-term asset and liability cash flow forecasting. However, with the assumption that future trends will follow recent historical trends, this study could also be applied for long-term actuarial use. In addition, we project future mortality rates for actuarial use with Chinese data. The CMI (Continuous Mortality Investigation) Mortality Projections Model developed by the Institute and Faculty of Actuaries is applied for modelling purposes. The model adopts a convergence structure from "initial" to "long-term" rates of mortality improvement as the process of projection. The initial rates of mortality improvement are derived using a 2D P-Spline methodology, and are then decomposed into age/period and cohort components. Given the short history of Chinese data, the long-term rates of mortality improvement are determined by borrowing information from international experience. K-means clustering with Dynamic Time Warping (DTW) distance is used to classify populations, which is novel in the actuarial mortality research field. The original CMI approach is deterministic, however, in this paper we incorporate stochastic elements using techniques outlined by Koller (2011) and described by Browne et al. (2009). Comparing our results with a pure extrapolative approach, we find that the modified CMI Mortality Projections Model is more suitable for long-term projections in China. Further, we conduct the first study of long-term age-sex-specific mortality forecasting for subpopulations in different areas of China: cities, towns and counties using the modified CMI Mortality Projections Model. From the historical experience, we find that people in cities have lower mortality rates and higher mortality improvement rates than people in towns and counties for most ages. If this trend continues, the mortality of different areas will diverge further in the future. From the projection results, we find that there will be significant mortality and life expectancy differences between cities, towns and counties for both males and females. By conducting sensitivity analysis, we also find that the life expectancy differences could be reduced by incorporating higher long-term mortality improvement rates for towns and counties, or increasing the speed of convergence from {u0300}initial' to {u0300}long-term' mortality improvement rates for males in counties. Uncertainties are attached to the central estimates to overcome the limitations of the original CMI approach from which only deterministic results can be obtained.
Description
Keywords
Citation
Collections
Source
Type
Thesis (PhD)
Book Title
Entity type
Access Statement
License Rights
Restricted until
Downloads
File
Description