Initial insurance losses are often reported with a textual description of the claim. The claims manager must determine the adequate case reserve for each known claim. In this paper, we present a framework for predicting the amount of loss given a textual description of the claim using a large number of words found in the descriptions. Prior work has focused on classifying insurance claims based on keywords selected by a human expert, whereas in this paper the focus is on loss amount prediction with automatic word selection. In order to transform words into numeric vectors, we use word cosine similarities and word embedding matrices. When we consider all unique words found in the training dataset and impose a generalised additive model to the resulting explanatory variables, the resulting design matrix is high dimensional. For this reason, we use a group lasso penalty to reduce the number of coefficients in the model. The scalable, analytical framework proposed provides for a parsimonious and interpretable model. Finally, we discuss the implications of the analysis, including how the framework may be used by an insurance company and how the interpretation of the covariates can lead to significant policy change. The code can be found in the TAGAM R package (github.com/scottmanski/TAGAM).

Piloting in the Panama Canal is exceptional as, due to its importance, the functions of the captains of vessels are taken over by pilots. Hence, prior to inauguration of the expanded canal, a limited number of pilots experienced on the existing canal were certified for the transit of Neopanamax vessels by means of planned and innovative individual learning. After this organisational training through operative training, with the implementation of the expanded canal in June 2016, the routine training started. Hence the learning curve in the performance of these manoeuvres will represent the growing skill acquired by both the pilots and the organisation. Given that the learning effect is measurable, this paper has the dual objective of determining two curve models: the organisation operative learning curve model and the routine learning curve model for pilots performing transit manoeuvres in the expanded Panama Canal waterways and the Cocolí and Agua Clara locks. Manoeuvre times in locks and transit in the whole of the canal were followed up continuously in the first 42 months of operation.

In this paper the catastrophe bond prices, as determined by the market, are analysed. The limited published work in this area has been carried out mainly by cat bond investors and is based either on intuition, or on simple linear regression on one factor or on comparisons of the process of cat bonds with similar features. In this paper a Generalised Additive Model is fitted to the market data. The statistical significance of different factors which may affect the cat bond prices is examined and the effect of these factors on the prices is measured. A statistical framework and analysis ould provide insight into the cat bond pricing and could have applications among other things in the construction of a cat bond portfolio, cat bond price indices and in understanding changes of the price of risk over time.

Despite many previous studies on the life history of the squid Loligo forbesi, the roles of internal and external factors in growth and maturation have not previously been investigated in detail. The present study takes advantage of the availability of statistical techniques, namely generalised additive models, which permit simultaneous evaluation of the effects of multiple explanatory variables while avoiding the need to assume linear relationships. This has the further advantage that the size of body parts can be entered directly into the models rather than expressed as indices (e.g., gonado-somatic index). The present analysis is based on measurements taken from monthly market samples during 1989-1991 and aims to infer patterns of reproductive and somatic investment though comparison of large numbers of individuals at different (post recruit) life-cycle stages. Results indicate that, once the effect of body size is removed, gonad weight in both sexes is affected by month and digestive gland weight, consistent with seasonal triggering of maturation, and with energy for gonad growth being derived from food. There is also evidence of a negative relationship between somatic and ovary growth in females, possibly indicating remobilisation of somatic tissue to grow the ovary. In males, relationships between variables tended to have more complex forms, probably related to the existence of two or more growth patterns leading to different sizes at maturity. As found in previous work, maturation begins around 1-2 months earlier in males, suggesting that sensitivity to external triggers is controlled by a sex-dependent internal factor.