Exploring consumer differences in food demand: a stochastic frontier approach

Abstract

Purpose The purpose of this paper is to consider a new application of stochastic frontier analysis, in which the method is applied to demand data for a food product category, in an attempt to benchmark category consumption and segment food consumers. Design/methodology/approach In a unified, two‐stage approach, a stochastic frontier model is first estimated and subsequently deviations from the demand frontier are regressed on customer characteristics. The method is illustrated in scanner panel data. Findings A frontier demand function estimated in scanner data of a frequently‐bought food category has significant and consistent parameters. Specific descriptor variables can explain excessive category demand and profile customers with considerable sales potential. Research limitations/implications More work is needed to generalise the usefulness of the proposed model in different food categories. Future research may employ alternative functional specifications and explanatory variables. Practical implications The empirical identification of salient characteristics improves consumer understanding and can assist in the design of data‐driven marketing action. Applied researchers can use marketing and demographic variables that are found in standard consumer panels to estimate frontier models. Originality/value The paper introduces stochastic frontier analysis as a means to determine consumer differences in food demand. This is an important area for retailers, producers and researchers.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 16 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.