Extracting clusters from aggregate panel data: A market segmentation study

作者：

Highlights：

• We propose a method that finds homogeneous trajectories in aggregate panel data.

• The SQP algorithm is used to maximize the log-likelihood function.

• The p-values are computed from the pseudo-inverse Hessian matrix.

• A case study shows the application of the method in market segmentation.

摘要

•We propose a method that finds homogeneous trajectories in aggregate panel data.•The SQP algorithm is used to maximize the log-likelihood function.•The p-values are computed from the pseudo-inverse Hessian matrix.•A case study shows the application of the method in market segmentation.

论文关键词：Sequential quadratic programing,Cluster analysis,Panel data,Market segmentation

论文评审过程：Received 5 August 2015, Revised 25 August 2016, Accepted 7 October 2016, Available online 9 November 2016, Version of Record 9 November 2016.

论文官网地址：https://doi.org/10.1016/j.amc.2016.10.012