Application of Data Quality Metrics in Insurance

This thesis concerns the determination and management of information quality in relation to the introduction of the European Union's Solvency II regulatory directive for the insurance industry. The first section provides an introduction to information quality theory; general information quality principles and the causes and effects of poor quality are presented, along with a more detailed look at various conceptual approaches to information quality, an overview of its most oft-identified dimensions, and principles of the construction of quality metrics. The second section concerns the general properties of the insurance industry and the nature of the Solvency II directive. Then, a more detailed description of Solvency II's demands on (re)insurance undertakings with regards to data and its quality is given, viz. the presence of a quality perspective in the directive, information needs arising from newly prescribed processes, and a duty to provide certain data to regulatory authorities as well as the general public. This is supplemented with an overview of analyses gauging the industry's level of readiness for Solvency II. In the third section, a system of metrics is suggested for measuring the quality of information used by an insurance undertaking to complete the reports (directed at the regulator or the public) mandated by Solvency II. A subset of these metrics are then applied to publicly available data (the annual report of a Czech Republic-based insurance company) to determine its degree of agreement with the public reports prescribed by Solvency II. Several other metrics are then applied to generated data sets to demonstrate their behaviour and peculiarities.