In insurance analytics, textual descriptions of claims are often discarded, because traditional empirical analyses require numeric descriptor variables. This paper demonstrates how textual data can be easily used in insurance analytics. Using the concept of word similarities, we illustrate how to extract variables from text and incorporate them into claims analyses using standard generalized linear model or generalized additive regression model. This procedure is applied to the Wisconsin Local Government Property Insurance Fund (LGPIF) data, in order to demonstrate how insurance claims management and risk mitigation procedures can be improved. We illustrate two applications. First, we show how the claims classification problem can be solved using textual information. Second, we analyze the relationship between risk metrics and the probability of large losses. We obtain good results for both applications, where short textual descriptions of insurance claims are used for the extraction of features.

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).

The Gamma model has been widely utilized in a variety of fields, including actuarial science, where it has important applications in insurance loss predictions. Meanwhile, high dimensional models and their applications have become more common in the statistics literature in recent years. The availability of such high dimensional models have allowed the analysis of non-traditional data, including those containing textual descriptions of the response. In the models used in such applications, the dispersion may be designed to be related to a set of covariates, as opposed to being a single fixed value for the entire population. Following this approach, we incorporate a group Lasso type penalty in both the dispersion and the mean parameterization for a Gamma model, and illustrate its use in a predictive analytics application in actuarial science. In particular, we apply the method to an insurance claim prediction problem involving textual data analysis methods. Simulations are conducted to illustrate the variable selection and model fitting performance of our method.

The presented research discusses general approaches to analyze and model healthcare data at the treatment level and at the store level. The paper consists of two parts: (1) a general analysis method for store-level product sales of an organization and (2) a treatment-level analysis method of healthcare expenditures. In the first part, our goal is to develop a modeling framework to help understand the factors influencing the sales volume of stores maintained by a healthcare organization. In the second part of the paper, we demonstrate a treatment-level approach to modeling healthcare expenditures. In this part, we aim to improve the operational-level management of a healthcare provider by predicting the total cost of medical services. From this perspective, treatment-level analyses of medical expenditures may help provide a micro-level approach to predicting the total amount of expenditures for a healthcare provider. We present a model for analyzing a specific type of medical data, which may arise commonly in a healthcare provider’s standardized database. We do this by using an extension of the frequency-severity approach to modeling insurance expenditures from the actuarial science literature.