A5067831001- Consumer Market Behavior and Pricing
- Innovation Diffusion and Forecasting
- Organizational Management and Leadership
- Economic and Environmental Valuation
- Customer churn and segmentation
- Consumer Retail Behavior Studies
- Firm Innovation and Growth
- Complex Systems and Time Series Analysis
- Forecasting Techniques and Applications
- Economics of Agriculture and Food Markets
- Decision-Making and Behavioral Economics
- Monetary Policy and Economic Impact
- Wine Industry and Tourism
- Customer Service Quality and Loyalty
- Economic Growth and Productivity
- Digital Platforms and Economics
- Supply Chain and Inventory Management
- Sensory Analysis and Statistical Methods
- Business Strategy and Innovation
- Spatial and Panel Data Analysis
- Optimal Experimental Design Methods
- Consumer Behavior in Brand Consumption and Identification
- Complex Network Analysis Techniques
- Statistical Methods and Bayesian Inference
- Auditing, Earnings Management, Governance
Erasmus University Rotterdam
2015-2024
Tinbergen Institute
2006-2023
The Econometric Society
2004
Curtin University
1989
An accurate prediction of what a customer will purchase next is paramount importance to successful online retailing. In practice, history data readily available make such predictions, sometimes complemented with characteristics. Given the large product assortments maintained by retailers, scalability method just as important its accuracy. We study two classes models that use predict buy next, i.e., novel approach uses latent Dirichlet allocation (LDA), and mixtures Dirichlet-Multinomials...
Abstract We introduce a multi‐level smooth transition model for panel of time series, which can be used to examine the presence common nonlinear business cycle features across many variables. The is positioned in between fully pooled model, imposes such features, and heterogeneous allows unrestricted nonlinearity. second‐stage linking parameters that determine timing switches regimes observable explanatory variables, thereby allowing lead–lag relationships members. discuss representation,...
In a collaborative study with major Dutch retailer, the authors assess short- and long-term effects of 25% item reduction on category sales. On an aggregate level, assortment can lead to substantive short-term sales losses but only weak negative effect. Short-term are caused mainly by fewer purchases former buyers delisted detergent items. However, results also show that attracts new buyers. These partially offset among The collection supplemental process data perceptions actual search time...
Abstract To study preferences, respondents to a survey are usually asked select their most preferred option from set. Preferences can be estimated more efficiently if rank all alternatives. When some unable perform the ranking task, using complete may lead substantial bias. We introduce model which endogenously describes capabilities of individuals. Estimated preferences based on this efficient when at least individuals able than one item, and they do not suffer biases due inabilities...
The authors put forth a sales response model to explain the differences in immediate and dynamic effects of promotional prices regular on sales. consists vector autoregression that is rewritten error correction format, which allows disentangle from effects. In second level model, price elasticities, cumulative elasticity, long-term elasticity are correlated with various brand-specific category-specific characteristics. applied seven years data weekly 100 different brands 25 product...
This article examines the global spillover of foreign product introductions and takeoffs on a focal country's time to takeoff, using novel data set penetration for eight high-tech products across 55 countries. It shows how clout, susceptibility influences, intercountry distances affect patterns. The authors find that takeoffs, but not introductions, accelerate takeoff. larger country, higher its economic wealth, more it exports, clout has in process. In contrast, poorer tourists receives,...
In this paper, we propose and operationalize a new method for optimizing shelf arrangements. We show that there are important dependencies between the layout of stock-keeping unit (SKU) sales marketing effectiveness. The importance these is further shown by substantive profit gains obtain with our proposed optimization approach. basis model standard equation explains using item-specific effect parameters intercepts. Hierarchical Bayes (HB) fashion, augment second layer relates to SKU...
A scalable model-based approach to gain insights in dynamic purchase behavior for large product assortments and customer bases.
The growing trend of networking in recent years has led to an increase number loyalty program partnerships, most notably multi-vendor programs (MVLP). In MVLP (as other types LPs), cardholders frequently receive promotional mailings intended sales at the participating vendors. This study examines individual vendor and joint (multiple vendors) promotions on performance five main vendors within MVLP. Findings indicate low responsiveness LP-induced promotions. may be improved if multiple...
This article examines cross-price promotional effects in a dynamic context. Among other things, we investigate whether previously established findings hold when consumer and competitive dynamics are taken into account. Five main influential (asymmetric price effect, neighborhood asymmetric share private label versus national brand asymmetry) appear jointly the second layer of pooled HB-VEC-VARX model, together with brand- category-specific variables. study tests relative importance these key...
Multivariate Logit models are convenient to describe multivariate correlated binary choices as they provide closed-form likelihood functions. However, the computation time required for calculating choice probabilities increases exponentially with number of choices, which makes maximum likelihood-based estimation infeasible when many considered. To solve this, we propose three novel methods: (i) stratified importance sampling, (ii) composite conditional (CCL), and (iii) generalized method...
Summary Consumer products and services can often be described as mixtures of ingredients. Examples are the mixture ingredients in a cocktail different components travel time (e.g. in-vehicle out-of-vehicle time) transportation setting. Choice experiments may help to determine how respondent's choice product or service is affected by combination In such experiments, individuals confronted with sets hypothetical they asked choose most preferred from each set. However, there no studies on...
Accurately predicting the next shopping basket of a customer is important for retailers, as it offers an opportunity to serve customers with personalized product recommendations or lists. The goal next-basket prediction predict coherent set products that will buy next, rather than just single product. However, if assortment retailer contains thousands products, number possible baskets becomes extremely large and most standard choice models can no longer be applied. Therefore, we propose use...
We propose a method to include seasonality in any diffusion model that has closed-form solution. The resulting captures way naturally matches the original model's pattern. assumes additional sales at seasonal peaks are drawn from previous or future periods. This implies pattern does not influence underlying is compared with alternative approaches through simulations and empirical examples. As alternatives, we consider standard Generalized Bass Model (GBM) basic Model, which ignores...
In estimating a multiple integral, it is known that Monte Carlo methods are more efficient than analytical techniques when the number of dimensions beyond seven. general, sample-mean method better hit-or-miss method. However, volume domain in high-dimensional space interest, usually preferred. It because difficulty generalizing for computation domain. This paper develops technique to make such generalization possible. The can be interpreted as volume-preserving transformation procedure. A...