国際会議7th International Conference on Econometrics and Statistics（EcoStat2024）で主席研究員村上大輔が学会発表いたします。
◆Title: A fast and flexible space-time varying coefficient model selection
◆Authors: Daisuke Murakami, Shinichiro Shirota, Seiji Kajita, Mami Kajita
◆Abstract:
Space-time varying coefficient (STVC) model attracts attention these days as a flexible tool to explore the spatio-temporal patterns in regression coefficients. However, the model tends to suffer from a difficulty in balancing computationally efficiency and model flexibility. To break the bottleneck, this study develops a fast and flexible STVC modeling method. For flexible modeling, we assume multiple processes in each varying coefficient, including purely spatial, purely temporal, and space-time interaction processes with/without time cyclicity. While consideration of multiple processes can be time consuming, we combine a pre-conditioning method and a model selection procedure, inspired by reluctant interaction modeling, to select/specify the latent space-time structure computationally efficiently. Monte Carlo experiments show that the proposed method outperforms alternatives in terms of the coefficient estimation accuracy and computational efficiency. Finally, the proposed method is applied to a crime analysis with sample size of 279,360, and confirmed that the proposed method provides reasonable varying coefficient estimates.