11h00 - 11h25
Stochastic Bilevel Models for Revenue Management.
During the last 30 years, Hotel Revenue Management research has not proposed and solved models that simultaneously consider inventory assignment, price, length of stay and uncertainty. The purpose of this paper is to develop a new model inspired in both Bilevel Pricing and Two-stage Stochastic models to allow managers to account with useful data for pricing decisions based on a better understanding of consumer's behaviour and market uncertainty.
The main contributions of this study are the following. First, we develop a mathematical model that addresses several elements, such as uncertainty, length of stay, quality of service, capacity, and groups' preferences, among others. Second, we propose a data generation process to overcome atypical and/or irrelevant cases. Finally, we develop exact and heuristic resolution methods, which allow us to provide useful results for decision making in the Hotel Industry.
11h25 - 11h50
American-Style Options in Gaussian Jump-Diffusion Models: Estimation and Evaluation
We propose a quasi-analytical approach for valuing American-style options in Gaussian jump-diffusion models that extend Merton's (1976) setting. Our approach is based on dynamic programming coupled with finite elements. We perform a numerical investigation that shows convergence and efficiency. We also address the model estimation and report an empirical investigation based on Apple.
11h50 - 12h15
Market Deployment Planning Optimization for Born Global Firms
In today’s fast global economy, entrepreneurs have a tendency to ever more to holistically design their born global business ventures in early stages of foundation. Embedded in the business design process is market deployment roadmap planning which is formulated as a vehicle for analyzing the timing of entry into markets along the planning horizon. We introduce an approach for market deployment roadmap planning, based on self-organizing maps for clustering markets and an optimization model that considers multi period strategic revenue and cost effects of market deployment decisions. We provide empirical results for a case study.
12h15 - 12h40
Do Some Modelling Frameworks Perform Better than Others by Design? A Case of Bankruptcy Prediction Models Investigated
In this research, we address two important questions; namely, do some modelling frameworks perform better than others by design? and to what extent the choice and/or the design of explanatory variables and their nature affect the performance of modelling frameworks? Answers to these generic research questions are devised for classification problems applied to corporate failure prediction. Six statistical and stochastic modelling frameworks are investigated along with several potential performance improvement mechanisms. The performance of these modelling frameworks and improvement mechanisms is tested within a multi-criteria framework. Our findings reveal that, although conceptually some modelling frameworks are supposed to perform better than others by design, from an application perspective, both the nature of information these framework are fed with and the packaging of such information make a difference in performance.