The continuous large and rapid advances in hardware and software allow advanced analytics to be applied to new problems in an increasing number of business problems. Additionally, the advances allow larger datasets to be analysed. PAM Analytics has carried out many analytics projects in a wide range of business sectors.
Some of the projects that PAM Analytics has carried out are described briefly below. The imputation and segmentation database marketing projects, the Christmas sales and passenger demand forecasting projects, and the manufacturing project are described in detail in the downloadable PDFs on the left-hand side of this page.
A logistic regression model to predict the likelihood of being granted temporary accommodation paid for by the Council as a result of either being homeless or in danger of being declared homeless. The aim of the model is to help the Council identify which approaches are likely to result in successful homelessness applications in order to target prevention services at those people who are most likely to actually become homeless and result in a cost to the Council.
- A Poisson regression model to predict the number of offences committed by youths when they are on non-custodial sentences given that they already had committed offences. The model is used by the Council to understand the factors - a range of background, behavioural, court order and offence data - that are likely to lead to high re-offence rates, and so help them prioritise their resources to those youths with the highest risk profiles.
- A model based on proximity analysis to identify people who are not currently foster carers but are potentially good prospects based on their similarities calculated from a range of data to current foster carers. The model is used by the Council to help meet the growing demand for foster carers by using a more structured and targeted approach than is used currently..
Regression models for calculating how and to what extent the reputations of major companies in the UK, the US and Europe influence their share prices and shareholder value.