A hybrid dynamic forecast model for analyzing celebrity endorsement effects on consumer attitudes

Abstract

This study investigates the time-varying effects of celebrity endorsements on consumer purchase attitudes toward promoted products using a novel dynamic hierarchical multi-attribute attitude forecast model. The induced direct and indirect effects via constructs of product attributes and net product value are then incorporated into the proposed conceptual model, which is formulated with a discrete-time nonlinear stochastic system. An empirical study of product categories of sport shoes and notebook computers demonstrates the feasibility of the proposed methodology. The analytical results demonstrate the capability of the proposed model to forecast consumer attitudes toward a promoted product and reveal the potential heterogeneity in patterns of attitude changes characterized by product attributes, price, and endorser performance as perceived by consumers. Furthermore, we infer that celebrity endorsement can significantly influence consumer purchase attitudes via both direct and indirect effects through product-attribute construct.