Margaret Farrugia

Although Canadians have overwhelmingly accepted electronic banking, 'in branch' or over-the-counter banking will not totally disappear. Banks are changing and developing to cater to customer demands. For that purpose, they need to maintain an understanding of customer demand in order to compete in the retail financial industry. Recognizing the travel behaviour of customers and their concentration in certain areas of the region is important. The challenge for banks and other large institutions is to utilize customer information to its fullest potential. Customer datasets are usually very large especially when attributes of transaction behaviour are included. Because of the size of customer datasets as well as the obligation to maintain customer privacy, aggregation of data is necessary. The aggregation of data should still allow for efficient interpretation of spatial patterns. The purpose of this report is to examine the effects of data aggregation on the patterns of customers' banking transaction behaviour. Transaction behaviour includes the selection of delivery mode, the frequency of transactions, the percentage of transactions for each type of delivery mode and the distances traveled to make those transactions. This study first assesses the transaction patterns at the individual (or most disaggregated level), and uses the results as a benchmark to compare the calculations of the data aggregated into varying scales and zones. This study assesses two types of modifiable areal units; irregular and regular zones, each at three levels of aggregation for accuracy against the results of individual data. The aggregation effects are measured in terms of the difference between the aggregated data and the individual data.