Funded Phd Research
A Model for Estimation of Anonymous Visits on Websites
Statistics Department; Faculty Advisers: Eric Bradlow (primary), Elea Feit, Shane Jensen
Anonymous website visits, which can make up a substantial share of total visits, are a lost opportunity for companies’ targeted marketing and also skew analyses based on data by identifying visitors only. This research develops a model to probabilistically assign web sessions to anonymous users, which retailers can leverage to better target advertisements, product offerings, and other content to consumers with and without identifications and more accurately track statistics such as number of unique visitors. The model will be applied to actual data from a large clothing retailer to improve customized product suggestions.
Novak, Julie, Elea McDonnell Feit, Shane Jensen and Eric Bradlow (2015), “Bayesian Imputation for Anonymous Visits in CRM Data,” Working Paper, https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2700347