The correct specification of spatial models, and especially the choice weights matrix, represents a crucial decision for researchers using georeferenced data. However, few guidelines exist on which matrix is most appropriate in certain cases. This paper therefore (1) studies sensitivity testing estimating models with different specifications (2) formulates recommendations applying regarding model selection specification. research based Monte Carlo simulations synthetic Resumen La correcta...

Customer retention has become a focal priority. However, the process of implementing an effective campaign is complex and dependent on firms’ ability to accurately identify both at-risk customers those worth retaining. Drawing empirical simulated data from two online retailers, we evaluate performance several parametric nonparametric churn prediction techniques, in order optimal modeling approach, context. Results show that under most circumstances (i.e., varying sample sizes, purchase...

Traditional approaches to managing customer churn have typically concentrated on those customers most likely defect. While accurately predicting probability is important, this metric alone does not sufficiently empower managers make optimal decisions. Hence, the current study focuses relationship between retention incentives and profit maximisation. Specifically, we improve existing management practices by: (1) allowing for heterogeneity in incentive redemption behaviour, (2) introducing...

Abstract In this article, we present a Bayesian spatial factor analysis model. We extend previous work on confirmatory by including geographically distributed latent variables and accounting for heterogeneity autocorrelation. The simulation study shows excellent recovery of the model parameters demonstrates consequences ignoring dependence. Specifically, find inefficiency in estimates score means bias corresponding covariance matrix. apply to Schwartz value priority data obtained from 5...

Resource exchange theory suggests service recovery compensation is optimal when it commensurate with what was lost (e.g., refund for overcharging). However, in practice, companies cannot always follow the theory-driven prescriptions, and complaint literature remains silent on how to best recover such suboptimal situations. This study takes a resource-based stance propose Mix&Match, framework tangible offers (refunds, redeliveries, or credits) optimize customer retention lifetime value...

This paper modifies the well-known conjectural variation model of Clarke and Davis (1982) by taking into account partial cross ownership (PCO) links among firms. It is shown that, unlike in no PCO case, link between firms’ price-cost margins degree tacit collusion nonlinear presence PCO. Thus, ignoring their would most likely lead to biased results due misspecification. The applied Japanese banking sector 2003. We find that banks compete a modest collusive environment, while neglecting gives...