Publicado
2025-08-02

Estimación bayesiana del parámetro de decepción en modelos de elección discreta bajo el paradigma RRM (Random Regret Minimization): aplicación al diseño de exhibiciones en grandes superficies

Bayesian Estimation of the Disappointment Parameter in Discrete Choice Models under the RRM (Random Regret Minimization) Paradigm: Application to Display Design in Large Retail Stores

DOI: https://doi.org/10.15332/23393076.11213
Carlos Gabriel Contreras Serrano

Resumen (es)

El presente estudio compara dos enfoques metodológicos para analizar experimentos de elección discreta con el fin de identificar la estrategia \'optima de organización de productos en góndolas de supermercados y reducir la complejidad cognitiva para el comprador durante el proceso de compra. Se contrastan el modelo Logit Multinomial Mixto basado en la Maximización de la Utilidad Aleatoria (RUM) y el Modelo Logit Multinomial Mixto con estimación bayesiana de parámetros bajo el enfoque de Minimización del Arrepentimiento Aleatorio (RRM). El experimento, diseñado de manera factorial fraccionada y realizado con la participación de 748 compradores colombianos en la categoría de crema dental, revela que el modelo RRM bayesiano ofrece predicciones más sensibles en comparación con el MNL RUM, especialmente en productos con participaciones de mercado medianas o pequeñas. Asimismo, la obtención de distribuciones posteriores de los parámetros que impactan variables como precio, beneficios, marca y posición en la góndola permite realizar simulaciones precisas de diversos escenarios. Estas simulaciones capturan la heterogeneidad no observada en las preferencias individuales de los compradores, lo que favorece la toma de decisiones en la negociación de estrategias de exhibición con grandes superficies. Ademós, las distribuciones posteriores ofrecen ventajas adicionales al contar con densidades empíricas conocidas, superando las limitaciones de los métodos estándar de estimación de parámetros en términos de optimización estadística.

Palabras clave (es): Discrete Choice Models, Mixed Multinomial Logit Model, RUM (Random Utility Maximization), RRM (Random Regret Minimization), Display strategy, Trade Marketing

Resumen (en)

This study compares two methodological approaches for analyzing discrete choice experiments aimed at identifying the optimal product organization strategy in supermarket shelves, while reducing cognitive complexity for buyers during the purchasing process. The study contrasts the Mixed Multinomial Logit model based on Random Utility Maximization (RUM) with the Bayesian parameter estimation approach of the Mixed Multinomial Logit model under the Random Regret Minimization (RRM) framework. The experiment, structured as a fractional factorial design and conducted with the participation of 748 Colombian shoppers in the toothpaste category, reveals that the Bayesian RRM model provides more sensitive predictions compared to the standard MNL RUM model, particularly for products with medium or small market shares. Additionally, the posterior distributions of parameters affecting variables such as price, benefits, brand, and shelf position enable precise simulations of various scenarios. These simulations capture unobserved heterogeneity in individual consumer preferences, facilitating decision-making in negotiations regarding shelf display strategies with major retailers. Moreover, the posterior distributions offer additional advantages by providing known empirical densities, overcoming the limitations of standard parameter estimation methods in terms of statistical optimization.

Referencias

Albert, J., & Chib, S. (1993). Bayesian analysis of binary and polychotomous response data. Journal of the American Statistical Association.

Allenby, G. (1997). An introduction to hierarchical Bayesian modeling. Advanced Research Techniques Forum. Tutorial Notes: American Marketing Association.

Allenby, G., & Rossi, P. (1999). Marketing models of consumer heterogeneity. Journal of Econometrics.

Ancarani, F., & Shankar, V. (2004). Price levels and price dispersion within and across multiple retailer types: Further evidence and extension. Journal of the Academy of Marketing Science.

Ato, M., López, J., & Benavente, A. (2013). Un sistema de clasificación de los diseños de investigación en psicología. Anales de Psicología.

Bauer, R., Menrad, K., & Decker, T. (2015). Adaptive hybrid methods for choice-based conjoint analysis: A comparative study. International Journal of Marketing Studies.

Bell, D. E. (1982). Regret in decision making under uncertainty. Operational Research.

Ben-Akiva, M., McFadden, D., Train, M., Walker, J., Bhat, C., Bierlaire, M., & Munizaga, M. (2002). Hybrid choice models: Progress and challenges. Marketing Letters.

Bhat, C. (2001). Quasi-random maximum simulated likelihood estimation of the mixed multinomial logit model. Transportation Research Part B: Methodological.

Bianchi-Aguiar, T., Hübner, A., Carravilla, M. A., & Oliveira, J. (2001). Retail shelf space planning problems: A comprehensive review and classification framework. European Journal of Operational Research.

Chorus, C. G. (2010). A new random regret minimization. European Journal of Transport Infrastructure.

Chorus, C. G. (2012). Random regret-based discrete choice modelling: A tutorial. Springer.

Contreras, G. (2020). Modelos econométricos de elección desde la economía del comportamiento: Modelamiento de elección discreta basada en costo emocional aleatorio-Aplicación a la industria agroquímica Colombiana. Comunicaciones en Estadística.

Farley, K., Thompson, C., Hanbury, A., & Chambers, C. (2013). Exploring the feasibility of conjoint analysis as a tool for prioritizing innovations for implementation. Implementation Science.

Gamerman, D. (1997). Sampling from the posterior distribution in generalized linear models. Statistics and Computing.

Greene, W. H., & Hensher, D. A. (2010). Modeling ordered choices: A primer. Cambridge University Press.

Hess, S., & Palma, D. (2019). Apollo: A flexible, powerful and customisable freeware package for choice model estimation and application. Journal of Choice Modelling.

Hübner, A., & Kuhn, H. (2024). Decision support for managing assortments, shelf space, and replenishment in retail. Flexible Services and Manufacturing Journal.

Lenk, P. (2009). Simulation of pseudo-bias correction to the harmonic mean estimator of integrated likelihoods. Journal of Computational and Graphical Statistics.

McCulloch, R., & Rossi, P. (1994). An exact likelihood analysis of the multinomial probit model. Journal of Econometrics.

McFadden, D., & Train, K. (2000). Mixed MNL models for discrete response. Journal of Applied Econometrics.

Muller, H., Kroll, E. B., & Vogt, B. (2012). Do real payments really matter? A re-examination of the compromise effect in hypothetical and binding choice settings. Marketing Letters.

Oppewal, H., Louviere, J., & Timmermans, H. (1994). Modeling hierarchical conjoint processes with integrated choice experiments. Journal of Marketing Research.

Pietrini, R., Paolanti, M., Mancini, A., Frontoni, E., & Zingaretti, P. (2024). Shelf management: A deep learning-based system for shelf visual monitoring. Expert Systems with Applications.

Revelt, D., & Train, K. (1998). Mixed logit with repeated choices: Households’ choices of appliance efficiency level. Review of Econometrics and Statistics.

Robert, C., & Casella, G. (2004). Monte Carlo statistical methods (2nd ed.). Springer.

Robinson, F., Feufel, M., Shalin, V., Steele-Johnson, D., & Springer, B. (2020). Rational adaptation: Contextual effects in medical decision making. Journal of Cognitive Engineering and Decision Making.

Rossi, P., Allenby, G., & McCulloch, R. (2005). Bayesian statistics and marketing. John Wiley & Sons.

Rossi, P., McCulloch, R., & Allenby, G. (1996). The value of purchase history data in target marketing. Marketing Science.

Shankar, V., Inman, J., Mantrala, M., Kelley, E., & Rizley, R. (2011). Innovations in shopper marketing: Current insights and future research issues. Journal of Retailing.

Silayoi, P., & Speece, M. (2007). The importance of packaging attributes: A conjoint analysis approach. European Journal of Marketing.

Simonson, I. (1992). The influence of anticipating regret and responsibility on purchasing decision. Journal of Consumer Research.

Tang, L., Luo, X., Cheng, Y., Yang, F., & Ran, B. (2014). Constructing an optimal orthogonal choice design with alternative-specific attributes for stated choice experiments. Transportation Research Record.

Train, K. (2001). A comparison of hierarchical Bayes and maximum simulated likelihood for mixed logit. Journal of Choice Modelling.

Train, K. (2009). Discrete choice methods with simulation. Cambridge University Press.

Train, K., & Sonnier, G. (2005). Mixed logit with bounded distributions of correlated partworths. Bayesian Analysis.

Dimensions

PlumX

Visitas

100

Descargas

Los datos de descarga aún no están disponibles.

Cómo citar

Contreras Serrano, C. G. (2025). Estimación bayesiana del parámetro de decepción en modelos de elección discreta bajo el paradigma RRM (Random Regret Minimization): aplicación al diseño de exhibiciones en grandes superficies. Comunicaciones En Estadística, 18(1). https://doi.org/10.15332/23393076.11213