{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/maximum-likelihood-estimation-of-a-finite","title":"Maximum likelihood estimation of a finite mixture of logistic regression models in a continuous data stream","arxiv_id":"1802.10529","date":"2018-02-28","proceeding":null,"authors":["Maurits Kaptein","Paul Ketelaar"],"abstract":"In marketing we are often confronted with a continuous stream of responses to\nmarketing messages. Such streaming data provide invaluable information\nregarding message effectiveness and segmentation. However, streaming data are\nhard to analyze using conventional methods: their high volume and the fact that\nthey are continuously augmented means that it takes considerable time to\nanalyze them. We propose a method for estimating a finite mixture of logistic\nregression models which can be used to cluster customers based on a continuous\nstream of responses. This method, which we coin oFMLR, allows segments to be\nidentified in data streams or extremely large static datasets. Contrary to\nblack box algorithms, oFMLR provides model estimates that are directly\ninterpretable. We first introduce oFMLR, explaining in passing general topics\nsuch as online estimation and the EM algorithm, making this paper a high level\noverview of possible methods of dealing with large data streams in marketing\npractice. Next, we discuss model convergence, identifiability, and relations to\nalternative, Bayesian, methods; we also identify more general issues that arise\nfrom dealing with continuously augmented data sets. Finally, we introduce the\noFMLR [R] package and evaluate the method by numerical simulation and by\nanalyzing a large customer clickstream dataset.","url_abs":"http://arxiv.org/abs/1802.10529v1","url_pdf":"http://arxiv.org/pdf/1802.10529v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"maximum-likelihood-estimation-of-a-finite","repo_url":"https://github.com/MKaptein/ofmlr","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null},{"paper_slug":"maximum-likelihood-estimation-of-a-finite","repo_url":"https://github.com/Nth-iteration-labs/ofmlr","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"marketing","task_name":"Marketing"},{"task_slug":"regression-1","task_name":"regression"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}