Finding market structure by sales count dynamics—Multivariate structural time series models with hierarchical structure for count data—

Terui, Nobuhiko; Ban, Masao; Maki, Toshihiro

doi:10.1007/s10463-009-0244-2

Cited by 12 publications

(23 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In contrast, Terui et al (2010) proposed a dynamic generalized linear model based on Poisson variables without over-dispersions. This represents a macromodel for aggregate sales directly without considering the microstructure.…”

Section: Models For Defining Market Structurementioning

confidence: 99%

“…Higher-order structure This model is extended to a higher-order hierarchical structure, as developed by Terui et al (2010). Next, we fully explain the model specifications, including the design matrix of state space priors, as this model is applied to actual time series data in the empirical analysis.…”

Section: Models For Defining Market Structurementioning

confidence: 99%

“…We use the store-level scanner, point of sales (POS), time series in the curry roux category that was applied to our previous model in Terui et al (2010) for comparison with the model with over-dispersion. The weekly series comprises three manufacturers that produce three products each, for a total of nine products during 110 weeks.…”

Section: Data and Variablesmentioning

confidence: 99%

“…Terui et al (2010) used this Poisson-multinomial relationship in a dynamic generalized linear model to propose a multivariate time series model with a hierarchical structure between variables for specifying market structure on the basis of a product's sales dynamics. Terui et al (2010) used this Poisson-multinomial relationship in a dynamic generalized linear model to propose a multivariate time series model with a hierarchical structure between variables for specifying market structure on the basis of a product's sales dynamics.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Multivariate Time Series Model with Hierarchical Structure for Over‐Dispersed Discrete Outcomes

Terui

Ban

2014

Journal of Forecasting

Self Cite

View full text Add to dashboard Cite

In this paper, we propose a multivariate time series model for over‐dispersed discrete data to explore the market structure based on sales count dynamics. We first discuss the microstructure to show that over‐dispersion is inherent in the modeling of market structure based on sales count data. The model is built on the likelihood function induced by decomposing sales count response variables according to products' competitiveness and conditioning on their sum of variables, and it augments them to higher levels by using the Poisson–multinomial relationship in a hierarchical way, represented as a tree structure for the market definition. State space priors are applied to the structured likelihood to develop dynamic generalized linear models for discrete outcomes. For the over‐dispersion problem, gamma compound Poisson variables for product sales counts and Dirichlet compound multinomial variables for their shares are connected in a hierarchical fashion. Instead of the density function of compound distributions, we propose a data augmentation approach for more efficient posterior computations in terms of the generated augmented variables, particularly for generating forecasts and predictive density. We present the empirical application using weekly product sales time series in a store to compare the proposed models accommodating over‐dispersion with alternative no over‐dispersed models by several model selection criteria, including in‐sample fit, out‐of‐sample forecasting errors and information criterion. The empirical results show that the proposed modeling works well for the over‐dispersed models based on compound Poisson variables and they provide improved results compared with models with no consideration of over‐dispersion. Copyright © 2014 John Wiley & Sons, Ltd.

show abstract

Section: Models For Defining Market Structurementioning

confidence: 99%

Section: Models For Defining Market Structurementioning

confidence: 99%

Section: Data and Variablesmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Multivariate Time Series Model with Hierarchical Structure for Over‐Dispersed Discrete Outcomes

Terui

Ban

2014

Journal of Forecasting

Self Cite

View full text Add to dashboard Cite

show abstract

“…Assuming f(t) is a general nonstationary time series, it accommodates a local trend when it exists with different levels over time. Terui, Ban, and Maki (2010) provide a recent application to a marketing problem. West and Harrison (1997) call this setting the "first-order (local linear) polynomial model" in their dynamic linear model.…”

Section: Preference Dynamicsmentioning

confidence: 99%

Dynamic Brand Satiation

Hasegawa

Terui

Allenby

2012

Journal of Marketing Research

Self Cite

View full text Add to dashboard Cite

The authors develop a dynamic factor model of brand satiation to explain longitudinal variation in consumer purchases. Factor loadings are associated with a brand's position along a satiation dimension, and factor scores are associated with a household's sensitivity to satiation effects. The authors introduce dynamics by allowing the factor scores to evolve over time, reflecting variation in household satiation sensitivity. They embed the factor model in a direct utility model that allows for both corner and interior solutions and show that it fits the data better than alternative specifications. Analysis of a panel data set of corn chips purchases indicates that respondent satiation is better explained by a low-dimensional factor structure, while baseline utility and preferences are not. The authors explore implications for product line assortment in the face of quickening satiation. Dynamic Brand SatiationConsumer tastes and preferences change over time for a variety of reasons; for example, preferences change with price, during promotional events, and after new product introductions. They also change with consumer learning when consumers grow tired of offerings, because of variety seeking and state-dependent behavior and because of changing sensitivities to product characteristics due to changing consumption contexts. A critical aspect in studying the dynamics of preference is determining whether changes are related to specific product attributes or combinations of attributes and which attributes lead to greater satiation in demand for offerings. If dynamic brand preferences are tied to product attributes, firms can consider physically reformulating brands to make them more attractive to consumers.Challenges arise in quantifying the dynamics of brand satiation because of the large number of brands and varieties in a category and an even larger number of attributes used to formulate the brands. Most brands comprise many * Shohei Hasegawa is a doctoral candidate (

show abstract

A review of Bayesian dynamic forecasting models: Applications in marketing

Migon

Alves

Menezes

et al. 2023

Appl Stoch Models Bus & Ind

View full text Add to dashboard Cite

We briefly review the main developments of Bayesian dynamic models. The emphasis is on marketing applications. Typical examples in this area are discussed. The concepts of monitoring and intervention are carefully explained with illustrative examples and open source computational routines. We avoid algebraic developments and instead use graphical examples to illustrate theoretical aspects. Two real‐world problems using Bayesian dynamic models are discussed. Finally, we describe recent developments and alternative proposals to formally address the dependence when dealing with the modeling of multiple time series.

show abstract

Finding market structure by sales count dynamics—Multivariate structural time series models with hierarchical structure for count data—

Abstract: Count data, Generalized linear model, Hierarchical market structure, MCMC, Poisson–multinomial distribution, Predictive density, POS time series,

Cited by 12 publications

References 13 publications

Multivariate Time Series Model with Hierarchical Structure for Over‐Dispersed Discrete Outcomes

Multivariate Time Series Model with Hierarchical Structure for Over‐Dispersed Discrete Outcomes

Dynamic Brand Satiation

A review of Bayesian dynamic forecasting models: Applications in marketing

Contact Info

Product

Resources

About