Yamada Hiroki

In Malaysia, the rapid proliferation of shopping malls has heightened consumer expectations, driving malls to diversify their scale, formats, and characteristics. This study employs block clustering to simultaneously classify consumers (rows) alongside area brand asset factors in Subang Jaya and store image factors at Sunway Pyramid (columns). Compared to traditional methods like K-means, this machine learning technique more efficiently processes complex data interrelationships. The analysis revealed minimal overlap between area-specific and mall-specific factors within the expectation clusters, suggesting a relatively weak correlation between the two. Furthermore, the identified consumer clusters reflect distinct trends in age, transportation, number of shopping companions, and the prioritization of the area versus the mall. These findings offer strategic insights into leveraging area brand assets and store image factors to enhance regional appeal and shopping mall satisfaction.

This study employs hierarchical Bayesian binomial logit models to examine the store image indicators that influence customer satisfaction at competing shopping malls in Malaysia, and whether area attractiveness and brand asset indicators affect shopping mall satisfaction. We construct proposed shopping mall satisfaction and area attractiveness ranking models for quantification with store image indicators and area attractiveness as explanatory variables, and the area attractiveness ranking model uses area brand asset indicators. The areas under assessment are Subang Jaya and Petaling Jaya. The shopping malls examined in this study are Sunway Pyramid, located in Subang Jaya, and 1Utama shopping center, located in Petaling Jaya. As the two areas are adjacent, the malls share a competitive relationship. We also estimate the two models simultaneously, providing measures to enhance shopping mall satisfaction and area attractiveness based on the estimation results of the models.

This study focuses on food supermarkets (hereinafter referred to as “supermarkets”). The objective is to use survey data to elucidate what perspectives (constituent concepts) consumers adopt when evaluating stores, depending on the set of stores under consideration (a single store versus two or more stores). Specifically, using survey data collected from consumers who shop in areas with intense supermarket competition, we will classify responses into those evaluating only one store and those evaluating two or more stores. We will then extract the perspectives (constituent concepts) that consumers prioritize when evaluating stores, and use confirmatory factor analysis to clarify the relationships between these perspectives and the observed variables, as well as the relationships among the constituent concepts themselves. Note that the stores consumers can evaluate are limited to those where they have made a purchase and plan to continue shopping in the future as part of their regular shopping routine. It is common for the set of products that consumers evaluate and consider to change before they reach the stage of purchasing a product. According to Moriguchi and Takemura (2018), the process leading to product selection involves a progression from the “accessible set”—the set of stores consumers can access—to the “known set”—stores consumers are aware of—to the “consideration set”—stores that could be potential choices—to the “selection set”—the final set of candidates—and finally to the “selection” itself. This study focuses specifically on the consideration set and aims to clarify the evaluation structures consumers possess based on differences in the consideration set.

This study aims to reveal the satisfaction structure of department stores located in specific urban regeneration urgent development areas and the degree of influence of the attractiveness of the areas. Specifically, we analyze the effects of store image factors and area attractiveness ranking on department store satisfaction ranking, and the effect of area brand asset factors on area attractiveness ranking simultaneously. We use a hierarchical Bayesian binomial logit model framework, and express the structure of department store satisfaction ranking and area attractiveness ranking for each consumer in the within-individual model. As a result of model estimation, we clarified store image factors that influence store satisfaction ranking, and area brand asset factors that influence area attractiveness ranking. Furthermore, we found that area attractiveness ranking also had a significant impact on store satisfaction ranking, and simultaneous analysis revealed area brand asset factors that can influence store satisfaction ranking. Based on these results, we were able to discuss perspectives on marketing strategies and public relations activities that department storeʼs management should take into account.

In this study, we focus on two competing urban drugstore chains in the Nagoya Station area—Matsumoto Kiyoshi (Matsudo City) and Amano Drug (Nagoya City)—to elucidate the structure of consumer satisfaction for each individual customer through modeling, and to derive insights regarding potential marketing strategies for these drugstores. We adopt a hierarchical Bayesian binary logit model, using the satisfaction ranking of the drugstores as the dependent variable and explaining it with store image components. Furthermore, assuming that the store image components of interest differ for each consumer, we incorporate the variable selection model used by Gilbride, Allenby, and Brazell (2006) in their brand choice model into the individual-level model. These models not only evaluate the magnitude of response parameters that differ by consumer but also capture the structural heterogeneity of the models across consumers. Furthermore, consumer attributes (gender, shopping frequency, and whether or not they had a pre-purchase intention) are used as explanatory variables in the hierarchical model. While Yamada (2023) evaluated the satisfaction structure using cross-sectional data, this study employs panel data to perform more reliable model estimation.

In this study, we evaluate the attractiveness of the two competing commercial districts—Nagoya Sakae and Nagoya Station—using a statistical model that accounts for consumer heterogeneity. Specifically, we adopt a hierarchical Bayesian binary logit model framework and use data obtained from a survey of consumers who shop in these commercial districts. In addition to the regional brand assets proposed by Sugano and Wakabayashi (2008)—namely, economic infrastructure assets, lifestyle assets, historical and cultural assets, natural assets, food culture assets, and community assets—this study’s questionnaire also included a new category: commercial assets. The commercial assets category asks whether attractive stores and entertainment facilities are located within the area. The collected survey results are applied to a hierarchical Bayesian binary logit model to identify, for each consumer, the regional brand asset factors that influence area selection. Furthermore, based on the model’s estimation results, this study discusses the marketing strategies that the area should implement.

This study aimed to elucidate customer satisfaction with specific drugstores and reveal successful marketing measures. We used a Bayesian hierarchical logit model that accounts for the heterogeneity of consumers. The proposed model incorporated the variable selection model proposed in the brand selection model of Gilbride, Allenby, and Brazell (2006). The variable selection model expresses structural heterogeneity of individual brand selection models. By incorporating the variable selection model into the logit model, we could express the structural heterogeneity of the satisfaction model for each customer in addition to the difference in reaction parameter values. Therefore, we were able to capture the factors that clearly affect satisfaction. We used questionnaire data about Vdrug and Genky, which are competing in Gifu. The questionnaire included items about convenience, product lineup, product quality, price, store atmosphere, clerk service, and satisfaction ranking. When a Bayesian hierarchical model is used in the marketing field, panel data are often used for estimating the model stably (Yamada and Sato, 2019). In this study, however, we applied cross-sectional data to the proposed model. Therefore, the results derived from the model estimation carry implications for retail practice.

This study elucidates the degree of the influence of marketing measures and product loyalty (ie, convenience, shopping, and specialty goods) on department store customers’ purchase amounts using a model formed by a Bayesian hierarchical regression model framework. In addition, we use a Tobit model to infer the latent characteristics of zero purchase amounts and estimate customers’ inertial purchasing behaviors by modeling loyalty variables. Direct mail (DM) sent to each customer is used as the marketing measure. To build a different purchase amount mechanism for each customer, we incorporate a variable selection model. The results of the empirical analysis provide information about strategies department stores can adopt to increase customers’ purchase amounts.

This study aims to elucidate the satisfaction structure of competing large department stores in the Nagoya area using a model that accounts for consumer heterogeneity. Furthermore, the retail mix attributes influencing satisfaction—which include the six attributes used by Yamada (2021) (assortment, price, events, facilities, sales staff, and convenience)—are supplemented with sales floors that provide entertainment (hereinafter referred to as “sales floors”) and used in the questionnaire survey. Many department stores eliminated sales floors and facilities that provided entertainment as the bubble economy came to an end and sales stagnated. However, because this strategy caused customers—particularly families—to turn away from department stores, it is said that stores are now returning to their previous entertainment-oriented models. This study also examines whether these entertainment-providing sales floors in department stores influence consumer satisfaction.

Drug stores have recently expanded in size and widened available product categories. Efficient store management requires accurate and rapid data analysis and in-depth understanding of customers' purchasing characteristics. In this study, we apply the block clustering method to point-of-sale (POS) data with identifications (IDs) in drug stores in Japan's Gifu region to examine the purchasing characteristics of product categories. Since this method can simultaneously evaluate customers and product categories, it is possible to investigate data more easily and quickly than other analytical methods used for customers and product categories. We construct blocks combining clusters of customer and product categories and identify the customers that stores should prioritize and specific product categories that should bolster marketing measures. This study proposes a method that retailers can employ when considering decisions on marketing activities such as product lineups, sales floor layouts, or pricing strategies.

ID-POS data records “who,” “when,” “what,” “how many,” and “how much” was purchased. By accumulating ID-linked POS data, it is possible to track customer purchasing behavior, and department stores can conduct multifaceted analyses of this data to enhance their marketing activities. However, although most department stores have implemented loyalty programs (FSPs), they are not effectively utilizing ID-linked POS data, and it must be said that their use of this data remains insufficient. An approach such as the one taken in this study—which classifies product categories based on simultaneous purchase trends and identifies customer groups with similar purchasing patterns—is effective for implementing marketing activities that provide personalized service to customers even in an environment where the number of sales associates has been reduced. Furthermore, in previous evaluations of department stores, the private sales system has received little attention and has rarely been the subject of research. In that sense, this study evaluates department stores from a new perspective.

The purpose of this paper is to elucidate how cultural facilities in Japanese department stores have attracted consumers and whether they can contribute to future marketing strategies. Drawing on Product Life Cycle (PLC) theory in the field of marketing research, we define the period from the early days of department stores to the present. For each stage of the PLC, we examine the roles and positioning of cultural facilities as documented in department store corporate histories and newspaper articles. We also conducted field research at the Matsuzakaya Nagoya store, which features a comprehensive array of cultural facilities, to explore the future prospects for such facilities. Furthermore, we discuss whether cultural facilities can contribute to department store business strategies, which are currently experiencing stagnation.

The purpose of this study is to use modeling to elucidate how habitual behavior (product loyalty and store loyalty) and marketing initiatives (direct mail and events) influence the formation of average transaction value among department store customers. By comparing visit frequency over a two-year period, customers are classified into those whose visit frequency increases, those whose visit frequency remains stable, and those whose visit frequency decreases, in order to identify trends in the degree of influence exerted by inertial behavior and marketing initiatives. A hierarchical Bayesian regression model is used for this analysis. Furthermore, for modeling inertial behavior, we draw on Guadagni and Little (1983); regarding direct mail as a marketing initiative, we use direct mail announcing products, points, and benefits as variables, and for events, we use regular events, special events, and clearance sales as variables. From the results of the empirical analysis, we will derive insights regarding the refinement of marketing initiatives that can increase the average transaction value of department store customers.

The purpose of this study is to reveal how marketing affects customers’ purchase amount and number of visits in Japanese department stores. We model purchase amounts by using a hierarchical Bayes regression model and number of visits by using a hierarchical Bayes Poisson regression model. Furthermore, we estimate the latent factor behind price as the purchase amount per month with a Type-1 Tobit model and the structural heterogeneity of each customer with a model for variable selection. Direct mail and events are used as marketing measures. The analytical results reveal marketing measures that raise customers’ final purchase amounts.

In this paper, POS data with IDs of drugstores is applied to the analysis method proposed by Govaert & Nadif (2010). By using the Poisson latent block model with snacks purchase history data at drugstores located in the Gifu region, we cluster both the customers and the snacks brands simultaneously, and grasp the snack brand group that the customer group with a high purchase frequency choose. Furthermore, based on the model estimation, we discuss effective marketing measures in drugstores. Although we analyze only snack category, it can be extended to other product categories and is an effective way to summarize Big Data.

We attempt to clarify the mechanism of the amount of money of purchase by each customer in a department store in Nagoya city using the Bayesian regression model. Specifically, we use two types of explanatory variables, “Direct mail”, and ”Events”, to research a relationship between the purchase amount and department store’s strategy. Direct mail means the number of mails that the department store uses to inform their customers about new items, privileges of card holders, and storewide events. We also add “frequency coming to the store” as explanatory variables. Additionally, we adopt type-1 Tobit model to presume the optimal value for the dependent variable, if the dependent variable in the regression model is zero. Furthermore, we examine the relation between department strategy and the customer’s attributes by incorporating customer’s attributes into the hierarchical model.

The purpose of this study is to clarify how the retail mix actually implemented in department store locations influences the visit behavior of existing customers. Specifically, we model the relationship between marketing variables unique to this retail format and customer visit behavior within the framework of a hierarchical Bayesian Poisson regression model. A distinctive feature of this study is that it also models the formation mechanisms of the explanatory variables—store loyalty, direct mail, and promotional events. The model estimation results confirmed that direct mail has the greatest influence on the frequency of department store visits. The results of the retail mix weight parameters revealed that, among direct mail pieces, customers place the highest value on those announcing points and benefits, and among promotional events, they place the highest value on mid-year gift-giving events.

The purpose of this paper is to reveal how the evaluation of store image between competitive supermarkets in the same retail area influences the evaluation as the most important store. We model with the framework of a hierarchical bayesian logit model，and use the data that evaluate the elements that consist of store image attributes by the Internet questionnaire survey. We include convenience of shopping, quality of the goods, assortment of the goods, price of the goods, crew's attitude for store image attributes. From their analysis, we discuss about the marketing measure that the evaluation of existing customers is developed.

The purpose of this paper is to reveal how the evaluation about the library located in Gifu City influences its satisfaction. Most of the user in the library are elderly persons and mothers with children, however there aren’t many youngsters. I model with the framework of a hierarchical bayes model, and evaluate the elements with questionnaire data by the students at Gifu Shotoku Gakuen University. From their analysis, I discuss about the library’s service quality that the satisfaction of young people is developed.

The purpose of this paper is to elucidate how the retail mix actually implemented in department store locations affects the average purchase amount per visit (average transaction value) of existing customers. We model this relationship using a hierarchical Bayesian Tobit regression framework, incorporating marketing variables specific to the department store industry—such as direct mail and promotional events—as explanatory variables. The results of this empirical study provide insights into marketing strategies for increasing the average transaction value of existing customers.

The purpose of this paper is to clarify how department stores’ marketing mix strategies influence customers’ average spending per visit (average transaction value). We model this relationship using a hierarchical Bayesian regression framework, with loyalty (inertial behavior), direct mail, and promotional events serving as explanatory variables in the individual-level model. The results of the empirical analysis provide insights into the refinement of marketing strategies from the perspective of retaining existing customers.

The purpose of this research is to reveal how the marketing mix strategy in the department store influences the customer's shopping trip behavior to the department store. The model represents the mechanism of frequency of coming to the stores with framework of an hierarchical Bayes Poisson regression model. The model includes three explanatory variables: “Store loyalty”, “Direct mail”, and “Events”. These variables are also modeled by functional form with parameters. As a result, we could confirm that direct mail is one of the marketing mix variables that have the most effect on frequency of coming to the store.

This study aims to model customer purchasing behavior in department stores in a way that accounts for consumer heterogeneity, elucidate the underlying mechanisms, and gain insights into efficient, personalized marketing strategies for customer retention. The purchasing behaviors to be examined are: (1) department store customers’ store selection behavior (customer satisfaction); (2) department store customers’ store visit behavior; and (3) the formation of average transaction value among department store customers. By elucidating these mechanisms, we aim to gain insights for fostering customer retention at both the attitudinal and behavioral levels.

We construct a binary logit model to elucidate the satisfaction structure of individual consumers with regard to department stores. In this model, we explain individual consumers’ satisfaction rankings (which we regard as pseudo-store choices) in terms of their expectations and performance evaluations of retail mix attributes. Furthermore, the structure of individual consumers’ expectations and evaluations of retail mix attributes (five attributes: merchandise, facilities, sales staff, convenience, and events) is also modeled using retail mix attribute elements (24 elements across the five attributes, such as the trendiness of merchandise). Customer demographic factors are used as explanatory variables in the hierarchical model.

Using POS data with ID's from a department store’s FSP, we will: 1) classify product categories based on customers’ co-purchase tendencies (forming subcategories); and 2) segment customers based on their purchasing preferences within these subcategories. We will employ factor analysis and cluster analysis as our analytical methods. Based on these results, we will explore ways to organize store layouts to facilitate easy navigation and consider directions for direct mail campaigns and sales strategies that take customer purchasing patterns into account.

This study clarifies the structure of consumer satisfaction with department stores by constructing a binary logit model. The model explains consumers’ satisfaction rankings (considered as pseudo-store choices) in terms of the difference between performance evaluations and expectations regarding the retail mix: (1) a SERVQUAL-type model, which explains this difference using “gain” (when the difference is positive) and “loss” (when the difference is negative); (2) A Prospect Theory-based model, and (3) a SERVPERF-type model which explains it solely through performance evaluation. This study compares and examines these models. The data used consists of survey responses regarding competing department stores in the Ginza district, and the models are estimated using the maximum likelihood method.