C38 - Multiple or Simultaneous Equation Models: Classification Methods; Cluster Analysis; Principal Components; Factor ModelsReturn

The aim of the study is to identity the main factors that affect claims amount paid by insurers in case of road accidents and to predict losses from valid third-party liability insurance (MTPLI) policies until their expiration. Such an assessment is essential to adequately cover MTPLI policies and ensure the sustainable development of insurance companies. The geography of the study covers the MTPLI market of Europe in the main areas, but a deeper analysis of the impact of various factors, interactions, and interrelationships in MTPLI product is focused on Latvian market data due to availability of high-quality primary data. The research is based on the analysis of primary Latvian MTPLI policies data of more than 128,000 road traffic accidents that have occurred during the time period from 2014 till 2020. Risk driver selection was performed based on the existing scientific studies and correlation analysis of the sample set. Both linear and nonlinear forms of relationships were used for modelling. A multivariate modeling was used to identify significant risk factors and to quantify their impact on loss of incidents. Statistical stability of the models was tested using chi-squared, t-tests and p-values. Validation of models calibrated where done using prediction errors measurements: mean square error (MSE), root mean squared error (RMSE), and mean absolute error (MAE) assessment both within sample and out of sample technics. The results indicated that the driver’s behavior (penalties and Bonus-Malus) as well as vehicle parameters (weight and age), had significant impacts on crash losses.

This article evaluates the production possibilities of the electricity sector in selected EU countries. The estimates for production functions are based on the financial data of individual companies in the selected sector. The analysis was based on a linearized version of the two-factor Cobb-Douglas production function, which was subsequently modified to compare productivity results by company size and country. The countries were selected based on the results of a cluster analysis. The cluster analysis was performed using aggregated data on the shares of energy sources in production in the electricity sector. The results show that companies from countries with a high share of renewables (such as Denmark) perform the worst in terms of total productivity. Furthermore, it was found that large companies have significantly higher productivity when compared to their smaller competitors.

The paper presents a procedure that helps in revealing topics hidden in large collections of textual documents (such as customer reviews) related to a certain group of products or services. Together with identification of the groups containing the topics the lists of important expressions is presented which helps in understanding what characterizes these aspects most typically from the semantic point of view. The procedure includes determining an appropriate number of groups representing the prevailing topics, partitioning the documents into a desired number of groups using clustering, extracting significant typical features of documents from each group with application of feature selection methods, and evaluating the outcomes with the assistance of a human expert. The results show that the presented approach, consisting mostly of automated steps, is able to separate and characterize the aspects of a certain product as discussed by the customers and be later useful, e.g., for handling customer complaints, designing promotional campaigns, or improving the products.