Yasemin Boztug

A combined approach for segment-specific market basket analysis

Abstract Market baskets arise from consumers’ shopping trips and include items from multiple categories that are frequently chosen interdependently from each other. Explanatory models of multicategory choice behavior explicitly allow for such category purchase dependencies. They typically estimate own and across-category effects of marketing-mix variables on purchase incidences for a predefined set of product categories. Because of analytical restrictions, however, multicategory choice models can only handle a small number of categories. Hence, for large retail assortments, the issue emerges of how to determine the composition of shopping baskets with a meaningful selection of categories. Traditionally, this is resolved by managerial intuition. In this article, we combine multicategory choice models with a data-driven approach for basket selection. The proposed procedure also accounts for customer heterogeneity and thus can serve as a viable tool for designing target marketing programs. A data compression step first derives a set of basket prototypes which are representative for classes of market baskets with internally more distinctive (complementary) cross-category interdependencies and are responsible for the segmentation of households. In a second step, segment-specific cross-category effects are estimated for suitably selected categories using a multivariate logistic modeling framework. In an empirical illustration, significant differences in cross-effects and price elasticities can be shown both across segments and compared to the aggregate model.

Estimation with the Nested Logit Model: Specifications and Software Particularities

The paper discusses the nested logit model for choices between a set of mutually exclusive alternatives (e.g. brand choice, strategy decisions, modes of transportation, etc.). Due to the ability of the nested logit model to allow and account for similarities between pairs of alternatives, the model has become very popular for the empirical analysis of choice decisions. However the fact that there are two different specifications of the nested logit model (with different outcomes) has not received adequate attention. The utility maximization nested logit (UMNL) model and the non-normalized nested logit (NNNL) model have different properties, influencing the estimation results in a different manner. This paper introduces distinct specifications of the nested logit model and indicates particularities arising from model estimation. The effects of using various software packages on the estimation results of a nested logit model are shown using simulated data sets for an artificial decision situation.

Modeling Joint Purchases with a Multivariate MNL Approach

Our research examines the hypothesis that products chosen on a shopping trip to a supermarket indicate the preference interdependencies of consumers between different products or brands. The bundle chosen on the trip can be regarded as an indicator of a global utility function. This function implies a cross-category dependence of brand choice behavior. We hypothesize that the global utility function related to a product bundle is the result of the marketing-mix of the underlying brands. The structure of the chosen products allows us to uncover the impact of certain marketing-mix variables and product bundle buying behavior.

A Market Basket Analysis Conducted with a Multivariate Logit Model

The following research is guided by the hypothesis that products chosen on a shopping trip in a supermarket can indicate the preference interdependencies between different products or brands. The bundle chosen on the trip can be regarded as the result of a global utility function. More specifically: the existence of such a function implies a cross-category dependence of brand choice behavior. It is hypothesized that the global utility function related to a product bundle results from the marketing-mix of the underlying brands. Several approaches exist to describe the choice of specific categories from a set of many alternatives. The models are discussed in brief; the multivariate logit approach is used to estimate a model with a German data set.

A Combined Approach for Segment-Specific Analysis of Market Basket Data

There are two main research traditions for analyzing market basket data that exist more or less independently from each other, namely exploratory and explanatory model types. Exploratory approaches are restricted to the task of discovering cross-category interrelationships and provide marketing managers with only very limited recommendations regarding decision making. The latter type of models mainly focus on estimating the effects of category-level marketing mix variables on purchase incidences assuming cross-category dependencies. We propose a procedure that combines these two modeling approaches in a novel two-stage procedure for analyzing cross-category effects based on shopping basket data: In a data compression step we first derive a set of market basket prototypes and generate segments of households with internally more distinctive (complementary) cross-category interdependencies. Utilizing the information on categories that are most responsible for prototype construction, segment-specific multivariate logistic models are estimated in a second step. Based on the data-driven way of basket construction, we can show significant differences in cross- effects and related price elasticities both across segments and compared to the global (segment-unspecific) model.

Seduced by the Label: How the Recommended Serving Size on Nutrition Labels Affects Food Sales

As consumers pay greater attention to nutrition content when choosing food, voluntary front-of-pack labels have become popular tools for food marketers. However, voluntary nutrition labels provide certain freedoms regarding the disclosed information, which can be exploited. A common strategy is to disclose nutrition values based on smaller recommended serving sizes, which presents the nutrition amounts favorably on the label. Problematically, consumers can misinterpret such information and draw biased conclusions regarding product healthiness. This study uses purchase data with 61 products from both healthy (yogurt) and unhealthy (cookies) categories to analyze how recommended serving sizes on nutrition labels affect food sales. In line with our predictions, sales increased after a label introduction in the healthy (but not in the unhealthy) category for products with smaller recommended serving sizes. Since the least healthy versions within the category tend to have smaller recommended serving sizes, nutrition labels can stimulate sales of unhealthier food.

Testing the Multinomial Logit Model

A general test to check the adequateness of a regression model against nonparametric alternatives is presented. This test procedure is then applied to the well known multinomial logit model and its power is considered in a simulation study. Finally, the multinomial logit model is tested for a real scanner panel data set. The null hypothesis that the multinomial logit model is an adequate model for the real data set is rejected.

Investigating the Competitive Assumption of Multinomial Logit Models of Brand Choice by Nonparametric Modeling

SummaryThe Multinomial Logit (MNL) model is still the only viable option to study nonlinear responsiveness of utility to covariates nonparametrically. This research investigates whether MNL structure of inter-brand competition is a reasonable assumption, so that when the utility function is estimated nonparametrically, the IIA assumption does not bias the result. For this purpose, the authors compare the performance of two comparable nonparametric choice models that differ in one aspect: one assumes MNL competitive structure and the other infers the pattern of brands» competition nonparametrically from data.

An empirical test of theories of price valuation using a semiparametric approach, reference prices, and accounting for heterogeneity

In this paper we estimate and empirically test different behavioral theories of consumer reference price formation. Two major theories are proposed to model the reference price reaction: assimilation contrast theory and prospect theory. We assume that different consumer segments will use different reference prices. The study builds on earlier research by Kalyanaram and Little (1994); however, in contrast to their work, we use parametric and semiparametric approaches to detect the structure of the underlying data sets. The different models are tested using a program module in GAUSS that was able to account for heterogeneity. The model types were calibrated by a simulation study. The calibrated modules were then used to analyze real market data.

Dual-process theory and consumer response to front-of-package nutrition label formats

Nutrition labeling literature yields fragmented results about the effect of front-of-package (FOP) nutrition label formats on healthy food choice. Specifically, it is unclear which type of nutrition label format is effective across different shopping situations. To address this gap, the present review investigates the available nutrition labeling literature through the prism of dual-process theory, which posits that decisions are made either quickly and automatically (system 1) or slowly and deliberately (system 2). A systematically performed review of nutrition labeling literature returned 59 papers that provide findings that can be explained according to dual-process theory. The findings of these studies suggest that the effectiveness of nutrition label formats is influenced by the consumers dominant processing system, which is a function of specific contexts and personal variables (eg, motivation, nutrition knowledge, time pressure, and depletion). Examination of reported findings through a situational processing perspective reveals that consumers might prefer different FOP nutrition label formats in different situations and can exhibit varying responses to the same label format across situations. This review offers several suggestions for policy makers and researchers to help improve current FOP nutrition label formats.