“…Some scholars used other econometric methods to study, such as the long short-term memory (LSTM) approach (Mele and Magazzino, 2020), machine learning approach (Magazzino et al, 2021;Magazzino and Mele, 2022), logarithmic mean division index (LMDI) decomposition method (Zhou et al, 2019), propensity score matching and differences-in-differences (PSM-DID) (Zhou et al, 2019), quantile regression method (Cheng et al, 2022;Liu et al, 2022), input-output method (Liu and Zhao, 2021), partial equilibrium model (Yu et al, 2022), autoregressive distributed lag models (ARDL) (Wang, 2022). And they found the influencing factors of carbon emissions containing economic growth (Ren et al, 2021;Mele and Magazzino, 2020;Magazzino et al, 2021;Magazzino and Mele, 2022), energy transition (Ren et al, 2021), energy consumption and use (Udemba et al, 2020;Magazzino, 2016), technological innovation and technological progress (Ahmed et al, 2016;Gu et al, 2020;Cheng et al, 2022;Liu et al, 2022), global value chain participation (Liu and Zhao, 2021), financing constraints (Yu et al, 2022), foreign direct investment (Zeng and Ye, 2019), and capital allocation efficiency (Zhao et al, 2021). Some scholars have also considered spatial correlation in studies of carbon emission, but most of them use SDM (Gu et al, 2020;Lv et al, 2019;Wu and Zhang, 2021;Xue et al, 2020;Zhang Y et al, 2018;Zhao et al, 2021;.…”