Data mining is an important new development in financial modelling - it helps companies manage, understand and get maximum value from their customer base.
As part of their daily business, companies collect and store enormous volumes of data regarding their operations and their customers. Traditional data analysis and modeling techniques are often unwieldy and time consuming in these situations or fail to pick up key relationships in the data. Data mining is a collection of techniques giving an efficient way of analysing these large databases in order to add value to any corporation.
PricewaterhouseCoopers Actuarial utilises in-house developed techniques as well as commercially available data mining packages to analyse the relationships and patterns that can be hidden within these databases.
A typical application of data mining is to identify which customers are more likely to take up a new product offering. This enables a more targeted advertising campaign and a consequent savings in distribution costs.
In the insurance setting, PricewaterhouseCoopers Actuarial has recently used data mining techniques to assist a workers’ compensation insurer to efficiently allocate claims and injury management resources by identifying which reported claims are most likely to result in long periods off work therefore high costs to the insurer.
Key specialist: Steven Lim
Relevant experience:
More than 10 years experience in the financial services sector
Provided financial modeling advice to various financial institutions including insurers, accident compensation schemes and banks
Applied data mining techniques in the insurance and banking sectors
Clients include CBA, Suncorp-Metway, Treasury Managed Fund, RACQ, RAA and RACTI
