<p>In the context of climate change, there is growing concern about the energy efficiency and indoor air pollution of residential buildings. The air exchange between indoors and outdoors for residential buildings plays a critical role in energy efficiency and indoor air quality, with the air exchange rate (AER) and air infiltration rate (AIR) being key parameters. However, existing research on residential AER and AIR is predominantly restricted to a few regions, and there is a significant lack of comprehensive global-scale analysis or predictive models for residential AER or AIR. This gap in data poses challenges in the estimation of energy consumption and indoor air quality globally. To address this issue, we conducted a comprehensive literature review to compile existing AER and AIR data worldwide. Recognizing the limitations in available data, we developed a Random Forest (RF) model, integrating key factors that influence AER and AIR, such as geographical location, climate parameters, building characteristics, and social development level, to analyze the collected data and estimate AER and AIR values across different regions and seasons globally. With the developed RF model, we generated the global seasonal AER and AIR data at a sub-national scale, addressed a critical data gap, and further provided a foundational dataset for global estimations or predictions of energy consumption and indoor air quality research in residential buildings.</p>