TY - THES UR - https://discovery.ucl.ac.uk/id/eprint/10197660/ TI - The Utility of Consumer Panels for Social and Geographic Research KW - Consumer Panel KW - Microsimulation KW - Machine Learning KW - Geocomputation KW - Gravity Models KW - Retail Geography N1 - Copyright © The Author 2024. Original content in this thesis is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0) Licence (https://creativecommons.org/licenses/by-nc/4.0/). Any third-party copyright material present remains the property of its respective owner(s) and is licensed under its existing terms. Access may initially be restricted at the author?s request. ID - discovery10197660 AV - restricted EP - 320 N2 - The overarching goal of this thesis is to evaluate the efficacy of consumer panels as an alternative data source for comprehending, profiling, and modelling the British Fast Moving Consumer Goods (FMCG) and Food On The Go (FOTG) markets. This research benefits from a collaboration with Kantar WorldPanel, utilising large-scale consumer panels designed to gather transaction data from nationally representative households across Great Britain. The contributions of the presented work are at three levels. First, an in-depth analysis is carried out concerning the characteristics, representativeness, and completeness of the provided datasets, critically outlining their utilities for data linkage with additional demand-side data from the addi- tional individual demographics panel and supply-side data Secondly, a contrast set mining model is developed to reveal insights into product-channel relationships in conjunction with regional and geodemographics groups. This process enables the detection of significant behavioural traits among population subgroups while providing further evidence how households moderate their product choices based on their retail alternatives and normative spatial patterns. Finally, the complexities underpinning grocery retailing are integrated and validated through the construction of a disaggregated Huff model. This model facilitates the understanding of store patronage decisions and their corresponding mobility patterns on a national scale. The conjunction of a large-scale geocoded consumer panel and disaggregated Huff models hold significant potential for modelling interactions between households and their surrounding retail establishments. It also underscores the theoretical jus- tification and extension for the development of retail gravity models at smaller area levels. Overall, the thesis aims to provide a robust evaluation of the benefits and limitations of utilising consumer panels for empirical social and geographic applications, and the methodological implications thereof. M1 - Doctoral PB - UCL (University College London) Y1 - 2024/09/28/ A1 - Tang, Jason ER -