Open AccessAssessing Direct Marketing Effectiveness at a Retailer’s: a Mixed-Effects Model of Heterogeneous Responsiveness to Personalized Promotions
Author(s) -
Roko Pedisic
Publication year - 2021
Publication title -
journal of statistical and econometric methods
Language(s) - English
DOI - 10.47260/jsem/1012
Subject(s) - coupon , random effects model , multilevel model , econometrics , mixed model , regression analysis , statistics , marketing , linear regression , mathematics , advertising , computer science , economics , business , medicine , meta analysis , finance
This study proposed a mixed linear modeling framework for revealing the averagetreatment effects of coupon promotions as well as for figuring out what kind ofcustomers are more responsive to coupon promotions. The analysis used detaileddata on 2,469 households who are frequent shoppers at a retailer over one year. Aseries of models starting from fixed parameters OLS regression to more flexiblerandom coefficient models were built. The results showed that receiving morecoupon promotions does not lead to higher spending among households. None ofthe available demographic characteristics was significantly linearly related to theresponsiveness to coupon promotions according the initial model with cross-levelinteraction terms. However, exploratory analysis of random slopes with the help ofa regression tree allowed to correct the specification, which allowed confirming thatpeople aged 25+ with annual household income exceeding $25,000 are moreresponsive, yet their response is insignificantly different from zero.Keywords: Mixed-Effects Model, Multilevel Analysis, HeterogeneousResponsiveness, Promotions.