Non-Melanoma Skin Cancer Mortality in Spain: A Predictive Model up to 2044

Non-melanoma skin cancers (NMSC) are the most common malignancies worldwide and are, worryingly, increasing in incidence. However, data in the literature on NMSC specific mortality are scarce, because these tumors are excluded from most mortality registries. The main objective of this study is to analyze NMSC’s mortality rates and use them to generate a predictive model for the coming years in Spain. Data on mid-year population and death certificates for the period 1979–2019 were obtained from the Spanish National Statistics Institute. The Nordpred program (Cancer Registry of Norway, Oslo, Norway) within statistical program R was used to calculate mortality adjusted rates, as well as the mortality projection with an age-period-cohort model. This is the first study to report a prediction about NMSC mortality in the next years. According to our findings, the number of NMSC deaths in older people will grow in both sexes, especially in those older than >85 years old (y.o.). The age-specific mortality rates of NMSC will tend to stabilize or gradually decrease, with the exception of women between 75–79 y.o., who will present a slight increase at the end of the period. Early prevention and screening of NMSC specifically oriented to this population might change this tendency.

A GROUP REGULARISATION APPROACH FOR CONSTRUCTING GENERALISED AGE-PERIOD-COHORT MORTALITY PROJECTION MODELS

Given the rapid reductions in human mortality observed over recent decades and the uncertainty associated with their future evolution, there have been a large number of mortality projection models proposed by actuaries and demographers in recent years. Many of these, however, suffer from being overly complex, thereby producing spurious forecasts, particularly over long horizons and for small, noisy data sets. In this paper, we exploit statistical learning tools, namely group regularisation and cross-validation, to provide a robust framework to construct discrete-time mortality models by automatically selecting the most appropriate functions to best describe and forecast particular data sets. Most importantly, this approach produces bespoke models using a trade-off between complexity (to draw as much insight as possible from limited data sets) and parsimony (to prevent over-fitting to noise), with this trade-off designed to have specific regard to the forecasting horizon of interest. This is illustrated using both empirical data from the Human Mortality Database and simulated data, using code that has been made available within a user-friendly open-source R package StMoMo.

Statistical features of persistence and long memory in mortality data

Understanding core statistical properties and data features in mortality data are fundamental to the development of machine learning methods for demographic and actuarial applications of mortality projection. The study of statistical features in such data forms the basis for classification, regression and forecasting tasks. In particular, the understanding of key statistical structure in such data can aid in improving accuracy in undertaking mortality projection and forecasting when constructing life tables. The ability to accurately forecast mortality is a critical aspect for the study of demography, life insurance product design and pricing, pension planning and insurance-based decision risk management. Though many stylised facts of mortality data have been discussed in the literature, we provide evidence for a novel statistical feature that is pervasive in mortality data at a national level that is as yet unexplored. In this regard, we demonstrate in this work a strong evidence for the existence of long memory features in mortality data, and second that such long memory structures display multifractality as a statistical feature that can act as a discriminator of mortality dynamics by age, gender and country. To achieve this, we first outline the way in which we choose to represent the persistence of long memory from an estimator perspective. We make a natural link between a class of long memory features and an attribute of stochastic processes based on fractional Brownian motion. This allows us to use well established estimators for the Hurst exponent to then robustly and accurately study the long memory features of mortality data. We then introduce to mortality analysis the notion from data science known as multifractality. This allows us to study the long memory persistence features of mortality data on different timescales. We demonstrate its accuracy for sample sizes commensurate with national-level age term structure historical mortality records. A series of synthetic studies as well a comprehensive analysis of real mortality death count data are studied in order to demonstrate the pervasiveness of long memory structures in mortality data, both mono-fractal and multifractal functional features are verified to be present as stylised facts of national-level mortality data for most countries and most age groups by gender. We conclude by demonstrating how such features can be used in kernel clustering and mortality model forecasting to improve these actuarial applications.

Mortality Projection on Indonesia's Abridged Life Table to Determine the EPV of Term Annuity

The life insurance industries usually use the Life Table for the valuation process, especially in calculating premiums and policy values of a policy. However, the Life Table is rarely updated; and it may even take years before they are updated. This happens because the insurers believe that the information in the Life Table is still related to the current state of a country and for the next several years. In fact, data and information related to mortality rates in a country are constantly changing and always being updated annually. Therefore, as an approach, researchers use the projection of mortality to approach the mortality rate in the future. Thus, future mortality data can be predicted so that better policies can be made by the governments or insurance industries. In this study, the Abridged Life Table of Indonesia is used in the projection of mortality for both sexes (male and female) of the population in Indonesia. The results of mortality projection are then used to calculate the Expected Present Value (EPV) of a term annuity-due under uniform distribution of deaths (UDD) for several values of and ages. The results obtained show that there is a decrease in the value of the mortality rate in the next few years. Therefore, it can be assumed that there is a possibility for longevity risk to occur in the future.

Bayesian Mixture Modelling for Mortality Projection

Although a large number of mortality projection models have been proposed in the literature, relatively little attention has been paid to a formal assessment of the effect of model uncertainty. In this paper, we construct a Bayesian framework for embedding more than one mortality projection model and utilise the finite mixture model concept to allow for the blending of model structures. Under this framework, the varying features of different model structures can be exploited jointly and coherently to have a more detailed description of the underlying mortality patterns. We show that the proposed Bayesian approach performs well in fitting and forecasting Japanese mortality.

A DOUBLE COMMON FACTOR MODEL FOR MORTALITY PROJECTION USING BEST-PERFORMANCE MORTALITY RATES AS REFERENCE

We construct a double common factor model for projecting the mortality of a population using as a reference the minimum death rate at each age among a large number of countries. In particular, the female and male minimum death rates, described as best-performance or best-practice rates, are first modelled by a common factor model structure with both common and sex-specific parameters. The differences between the death rates of the population under study and the best-performance rates are then modelled by another common factor model structure. An important result of using our proposed model is that the projected death rates of the population being considered are coherent with the projected best-performance rates in the long term, the latter of which serves as a very useful reference for the projection based on the collective experience of multiple countries. Our out-of-sample analysis shows that the new model has potential to outperform some conventional approaches in mortality projection.