Global Household Size and Energy Use Intensity Under Shared Socioeconomic Pathways from 1960 to 2100

This dataset includes synthetic household sizes and end-use energy intensities for countries around the world, spanning various Shared Socioeconomic Pathways (SSP) from 1960 to 2100. Historical data up to 2021 was utilized to calibrate an XGBoost algorithm, which allows for interpolation and extrapolation of data in one-year intervals up to 2021 and in five-year intervals between 2025 and 2100. Five SSPs were predicted: SSP1, SSP2, SSP3, SSP4, and SSP5.Household sizes were predicted for each country, covering the national average, urban areas, and rural regions. Harmonized household sizes from the CORESIDENCE project (https://www.nature.com/articles/s41597-024-02964-3) were used to train an XGBoost model with customized objective function and evaluation metrics to prevent rural and urban household sizes from being both greater or smaller than the national average.Each country's end-use energy intensity for cooking, lights, appliances, and water heating was predicted using an XGBoost model. The free version of the IEA's Energy End-Uses and Efficiency Indicators (https://www.iea.org/data-and-statistics/data-product/energy-efficiency-indicators-highlights) was used.The model's predictors were obtained from the World Bank Database (https://data.worldbank.org), and future projections for each SSP of the same predictors were obtained from the Wittgenstein Centre Human Capital Data Explorer (https://dataexplorer.wittgensteincentre.org/wcde-v3/).