Actuaries must model mortality to understand, manage and price risk. Continuous-time methods offer considerable practical benefits to actuaries analysing portfolio mortality experience. This paper discusses six categories of advantage: (i) reflecting the reality of data produced by everyday business practices, (ii) modelling rapid changes in risk, (iii) modelling time- and duration-varying risk, (iv) competing risks, (v) data-quality checking and (vi) management information. Specific examples are given where continuous-time models are more useful in practice than discrete-time models.

We reprise some common statistical models for actuarial mortality analysis using grouped counts. We then discuss the benefits of building mortality models from the most elementary items. This has two facets. First, models are better based on the mortality of individuals, rather than groups. Second, models are better defined in continuous time, rather than over fixed intervals like a year. We show how Poisson-like likelihoods at the “macro” level are built up by product integration of sequences of infinitesimal Bernoulli trials at the “micro” level. Observed data is represented through a stochastic mortality hazard rate, and counting processes provide the natural notation for left-truncated and right-censored actuarial data, individual or age-grouped. Together these explain the “pseudo-Poisson” behaviour of survival model likelihoods.

We consider pricing of a specialised critical illness and life insurance contract for breast cancer (BC) risk. We compare (a) an industry-based Markov model with (b) a recently developed semi-Markov model, which accounts for unobserved BC cases and progression through clinical stages of BC, and (c) an alternative Markov model derived from (b). All models are calibrated using population data in England and data from the medical literature. We show that the semi-Markov model aligns best with empirical evidence. We then consider net premiums of specialized life insurance products under various scenarios of cancer diagnosis and treatment. The results show strong dependence on the time spent with diagnosed or undiagnosed pre-metastatic BC. This proves to be significant for refining cancer survival estimates and accurately estimating related age dependence by cancer stage. In contrast, the industry-based model, by overlooking this critical factor, is more sensitive to the model assumptions, underscoring its limitations in cancer estimates.

In this paper, we construct interpretable zero-inflated neural network models for modeling hospital admission counts related to respiratory diseases among a health-insured population and their dependants in the United States. In particular, we exemplify our approach by considering the zero-inflated Poisson neural network (ZIPNN), and we follow the combined actuarial neural network (CANN) approach for developing zero-inflated combined actuarial neural network (ZIPCANN) models for modeling admission rates, which can accommodate the excess zero nature of admission counts data. Furthermore, we adopt the LocalGLMnet approach (Richman & Wüthrich (2023). Scandinavian Actuarial Journal, 2023(1), 71–95.) for interpreting the ZIPNN model results. This facilitates the analysis of the impact of a number of socio-demographic factors on the admission rates related to respiratory disease while benefiting from an improved predictive performance. The real-life utility of the methodologies developed as part of this work lies in the fact that they facilitate accurate rate setting, in addition to offering the potential to inform health interventions.

Anti-selection occurs when information asymmetry exists between insurers and applicants. When an applicant knows they are at high risk of loss, but the insurer does not, the applicant may try to use this knowledge differential to secure insurance at a lower premium that does not match risk.