Download Credit Scoring, Response Modeling, and Insurance Rating: A by Steven Finlay (auth.) PDF
By Steven Finlay (auth.)
Read or Download Credit Scoring, Response Modeling, and Insurance Rating: A Practical Guide to Forecasting Consumer Behavior PDF
Best banking books
The appliance of knowledge Mining (DM) applied sciences has proven an explosive development in increasingly more diversified components of commercial, govt and technology. of an important enterprise parts are finance, specifically in banks and insurance firms, and e-business, reminiscent of internet portals, e-commerce and advert administration companies.
Every one new bankruptcy of the second one variation covers a facet of the fastened source of revenue marketplace that has develop into suitable to traders yet isn't really lined at a sophisticated point in latest textbooks. this is often fabric that's pertinent to the funding judgements yet isn't freely to be had to these no longer originating the goods.
Additional info for Credit Scoring, Response Modeling, and Insurance Rating: A Practical Guide to Forecasting Consumer Behavior
It must be possible to provide a clear and objective explanation as to why a model generates a given score in terms of the predictor variables. Simple models also aid understanding with a wider audience. Complex modeling solutions mean that understanding is restricted to a few specialist individuals. This makes it almost impossible for senior managers to make informed decisions about the validity or appropriateness of the models that have been constructed by the people who work for them. It is interesting to note that one of the conclusions 22 Credit Scoring, Response Modeling, and Insurance Rating of the Turner Report into the causes of the worldwide banking crisis of 2007–8 was that: The very complexity of the mathematics used to measure and manage risk, moreover, made it increasingly difﬁcult for top management and boards to assess and exercise judgement over the risks being taken.
Consider an insurer that has seen the number of new policies it underwrites falling by 2–3 percent year on year. At the start of the ﬁnancial year it has been agreed that to reverse the declining trend the company’s response and conversion models will be redeveloped – the goal being to generate more responders from the same marketing spend, and of these responders convert a greater proportion of them into policy holders. There are two proposals on the table. • Proposal 1. Construct a suite of new models, and at the same time upgrade the organizations systems to provide access to a range of new data sources that will mean far more predictive models can be constructed.
However, as we shall discuss in more detail later, what one means by “good payer” and “bad payer” is open to debate. One extreme is to consider someone a good payer only if they have never missed a payment over the entire term of the credit agreement. At the other end of the scale everyone may be considered a good payer unless they have missed so many payments that their loan has had to be written off.