This paper explores the impact of digital economy, population, affluence, technology, and other factors on carbon emissions, with panel data for 13 cities in the Beijing-Tianjin-Hebei urban agglomeration from 2011-2019. To overcome the negative influences of multicollinearity among independent variables under acceptable bias, we extended the traditional STIRPAT model and adopted the Partial Least Squares Regression (PLSR) algorithm. Results show that the digital economy has a directly dampening effect on carbon emissions, and the effect will diminish as the digital economy develops. Besides, under different development levels, differences are significant in terms of the impact of population, affluence, technology, urbanization rate and industrial structure on carbon emissions. Academically, we applied the PLSR method to the study of the relationship between digital economy and carbon emissions for the first time, which enhanced the credibility of the research conclusions. In addition, analysis based on samples from China's Beijing-Tianjin-Hebei urban agglomeration also provides more empirical evidence for related research. Practically, we recommend such policies as developing digital cities, promoting low-carbon concepts, and accelerating industrial transformation for the Beijing-Tianjin- Hebei region to achieve the "dual-carbon" goals and high-quality economic development.