Data and Program Code for "Nonlinear Temperature Sensitivity of Residential Electricity Demand: Evidence from a Distributional Regression Approach"
Citation: Nam, K., & Seo, W. K. (2026). Nonlinear temperature sensitivity of residential electricity demand: Evidence from a distributional regression approach. Energy Economics, 153: 109076.
@ Program Code Descriptions for “Nonlinear Temperature Sensitivity of Residential Electricity Demand: Evidence from a Distributional Regression Approach” October 2025 Kyungsik Nam
@ Main analysis files
The study is implemented with five MATLAB .m files, each aligned to a modeling specification and set of empirical results.
-
Main_FRKET_Density_Shock.m — Implements the density-to-demand model under heat and cold shock scenarios; generates results reported in Figures 1, 2, 3, 4, 6, and 8.
-
Main_FRKET_Hazard_Heat_Shock.m — Implements the hazard-to-demand model for heat-wave scenarios; corresponds to Figures 5 and 8.
-
Main_FRKET_Hazard_Cold_Shock.m — Implements the hazard-to-demand model for cold-wave scenarios; corresponds to Figures 5 and 8.
-
Main_FRKET_All_TRF.m — Implements the density- and hazard-to-demand models to produce benchmark TRFs (Figure 7) and cross-validation statistics (CV_Final, Table 2).
-
Main_FRKET_QtoD_Anal.m — Implements the quantile-to-demand model; corresponds to Figures 9–10.
@ Data files
-
Elec_demand_data.mat — Monthly residential electricity demand for Korea, January 1999–December 2023.
-
KTemp_Density.mat — Estimated temperature density functions for the same period, used as inputs to the distributional regression framework.
-
Temp_Quant_v202501.mat — Estimated temperature quantile functions for the same period, used as inputs to the distributional regression framework.
@ Inquiries For questions regarding the program code or data, please contact: Kyungsik Nam (ksnam@hufs.ac.kr) Division of Climate Change, Hankuk University of Foreign Studies