ABSTRACT
Market segmentation is an often-used marketing model for efficient allocation of a company's resources. It divides a population of customers into subpopulations—segments of customers. Customers within a segment are similar in their products and services, and customers across segments are dissimilar in their products and services. Market segmentation model implementation allows for effectively applying resources by targeting customers within their assigned segments. This chapter presents a novel approach to build a market segmentation model based on time-series data using latent class analysis (LCA). It describes k-means clustering and reviews principal component analysis (PCA). K-means clustering generates k mutually exclusive groups by distances computed from one or more quantitative variables (Xs). The fundamental underlying difference between LCA and k-means methodologies is that LCA is model-based, whereas k-means is a heuristic. A heuristic is any approach to solving a problem that uses intuition based on the problem domain.
