{"title":"Interdependences of Products in Market Baskets: Comparing the Conditional Restricted Boltzmann Machine to the Multivariate Logit Model","authors":"H. Hruschka","doi":"10.1515/roms-2020-0074","DOIUrl":null,"url":null,"abstract":"Abstract We analyze market baskets of individual households in two consumer durables categories (music, computer related products) by the multivariate logit (MVL) model, its finite mixture extension (FM-MVL) and the conditional restricted Boltzmann machine (CRBM). The CRBM attains a vastly better out-of-sample performance than MVL and FM-MVL models. Based on simulation-based likelihood ratio tests we prefer the CRBM to the FM-MVL model. To interpret hidden variables of conditional Boltzmann machines we look at their average probability differences between purchase and non-purchases of any sub-category across all baskets. To measure interdependences we compute cross effects between sub-categories for the best performing FM-MVL model and CRBM. In both product categories the CRBM indicates more or higher positive cross effects than the FM-MVL model. Finally, we suggest appropriate future research based on larger and more detailed data sets.","PeriodicalId":35829,"journal":{"name":"Review of Marketing Science","volume":"19 1","pages":"33 - 51"},"PeriodicalIF":0.0000,"publicationDate":"2020-11-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1515/roms-2020-0074","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Review of Marketing Science","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1515/roms-2020-0074","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"Business, Management and Accounting","Score":null,"Total":0}
引用次数: 0
Abstract
Abstract We analyze market baskets of individual households in two consumer durables categories (music, computer related products) by the multivariate logit (MVL) model, its finite mixture extension (FM-MVL) and the conditional restricted Boltzmann machine (CRBM). The CRBM attains a vastly better out-of-sample performance than MVL and FM-MVL models. Based on simulation-based likelihood ratio tests we prefer the CRBM to the FM-MVL model. To interpret hidden variables of conditional Boltzmann machines we look at their average probability differences between purchase and non-purchases of any sub-category across all baskets. To measure interdependences we compute cross effects between sub-categories for the best performing FM-MVL model and CRBM. In both product categories the CRBM indicates more or higher positive cross effects than the FM-MVL model. Finally, we suggest appropriate future research based on larger and more detailed data sets.
期刊介绍:
The Review of Marketing Science (ROMS) is a peer-reviewed electronic-only journal whose mission is twofold: wide and rapid dissemination of the latest research in marketing, and one-stop review of important marketing research across the field, past and present. Unlike most marketing journals, ROMS is able to publish peer-reviewed articles immediately thanks to its electronic format. Electronic publication is designed to ensure speedy publication. It works in a very novel and simple way. An issue of ROMS opens and then closes after a year. All papers accepted during the year are part of the issue, and appear as soon as they are accepted. Combined with the rapid peer review process, this makes for quick dissemination.