{"title":"品味会计","authors":"C.P.F. de Nijs","doi":"10.4324/9781315205922-13","DOIUrl":null,"url":null,"abstract":"Despite the abundance of advertisement and price promotion in marketing, consumption choices often rely just as heavily on personal taste. We compare the conventional Multinomial Logit discrete choice model, with a Mixed Logit specification, as to ascertain whether consumers behave hetero- or homogeneously. We consider both Multivariate Normal and discrete distributions for model parameters. Model performance is measured by in-sample information criteria, as well as out-of-sample forecasting performance. We find strong evidence of preference heterogeneity in both a product marketing setting, as well as a discrete choice survey experiment. By incorporating a Guadagni and Little (1983) loyalty variable, we extend this result to a dynamic choice setting, finding residual preference heterogeneity, even when brand loyalty is accounted for.","PeriodicalId":245295,"journal":{"name":"Taste, Consumption, and Markets","volume":"117 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2018-09-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Accounting for Taste\",\"authors\":\"C.P.F. de Nijs\",\"doi\":\"10.4324/9781315205922-13\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Despite the abundance of advertisement and price promotion in marketing, consumption choices often rely just as heavily on personal taste. We compare the conventional Multinomial Logit discrete choice model, with a Mixed Logit specification, as to ascertain whether consumers behave hetero- or homogeneously. We consider both Multivariate Normal and discrete distributions for model parameters. Model performance is measured by in-sample information criteria, as well as out-of-sample forecasting performance. We find strong evidence of preference heterogeneity in both a product marketing setting, as well as a discrete choice survey experiment. By incorporating a Guadagni and Little (1983) loyalty variable, we extend this result to a dynamic choice setting, finding residual preference heterogeneity, even when brand loyalty is accounted for.\",\"PeriodicalId\":245295,\"journal\":{\"name\":\"Taste, Consumption, and Markets\",\"volume\":\"117 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2018-09-21\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Taste, Consumption, and Markets\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.4324/9781315205922-13\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Taste, Consumption, and Markets","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.4324/9781315205922-13","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Despite the abundance of advertisement and price promotion in marketing, consumption choices often rely just as heavily on personal taste. We compare the conventional Multinomial Logit discrete choice model, with a Mixed Logit specification, as to ascertain whether consumers behave hetero- or homogeneously. We consider both Multivariate Normal and discrete distributions for model parameters. Model performance is measured by in-sample information criteria, as well as out-of-sample forecasting performance. We find strong evidence of preference heterogeneity in both a product marketing setting, as well as a discrete choice survey experiment. By incorporating a Guadagni and Little (1983) loyalty variable, we extend this result to a dynamic choice setting, finding residual preference heterogeneity, even when brand loyalty is accounted for.