Bert Kramer

N.E.W.S.: a model for the evaluation of non-life insurance companies

Abstract This paper describes a model which can be used by the Dutch insurance supervisor to determine the priority a non-life insurer should have for further examination. This model combines a traditional statistical technique (an ordered logit model) with artificial intelligence techniques (a neural network and an expert system). The output of the model consists of the priority for further examination (high, medium, or low), and a report which summarizes the main findings of the model. The model was able to adequately classify 93% of the companies in a 1993 data set.

An Asset Liability Management Model for Housing Associations

With asset liability management (ALM), all the relevant asset and liability classes are managed in an integrated fashion. We describe an ALM model for housing associations. This model uses simulation to show the development of a housing association, usually measured as solvency and profitability, dependent on both internal (strategy) and external (economy) factors. In order to assess the associations’ risk and return profile, we generate a large number of economic scenarios. Furthermore, we will show the pitfalls of just using one or a few scenarios. Finally, we will show how this model can be used to obtain insight into the influence and effectiveness of specific instruments.

A practice based methodology for comparing social objectives in housing associations

For housing associations, social objectives and financial restrictions will often be competing. Therefore, an important question is how a housing association should allocate its resources in order to achieve its goals. Traditionally, housing associations set financial targets and discuss social aspects, but they fail to set concrete targets for their social objectives which can be measured and evaluated properly. The problem is that social objectives are often difficult to quantify and compare. This paper introduces a method that makes it possible to compare social objectives and to actively involve social objectives in the decision-making process of housing associations. In this way, the decision-making process becomes more transparent and housing associations can optimise their social performance given the available capital.

A neural network model for the evaluation of Dutch non-life insurance companies

Based on six financial ratios, a one-hidden-layer back-propagation neural network classifies Dutch non-life insurance companies as strong moderate, or weak. The network shows very good performance for weak and strong companies (95% correct), but completely fails to recognize moderate companies. The relative importance of each input variable is analyzed by calculating the strength of the relationship between each input and each output variable.

Solvency-Simulation in Non-life Insurance

This paper describes the development of a cash flow model that can be used to analyze the solvency-position of a non-life insurance company. The model can be used by the management of a company to analyze the consequences of different policies on the solvency-position in the coming years. The model is build up of 3 modules: a financial module where the value of and the profits on the investment portfolio are determined, an insurance-technical module where the loss-amount is determined, and the main model where the results of the first two modules are joined into a balance-sheet from which the solvency margin can be calculated. The model uses simulation to generate the needed values in the financial and insurance-technical modules. This research is partly based on and inspired by the management model of Daykin and Hey (see Daykin & Hey (1990)).1