Chen, LiangDolado, Juan JoséGonzalo, JesúsRamos Ramirez, Andrey DavidUniversidad Carlos III de Madrid. Departamento de Estadística2022-07-222022-07-222013-07-222340-5031https://hdl.handle.net/10016/35531The relationship between global warming and CO2 is a long-standing question in theclimate change literature. In this paper we revisit this topic through the lenses of a new class of factor models for high-dimensional panel data, labeled Quantile Factor Models (QFM). This technique allows us to extract quantile-dependent factors from the distributions of changes in temperatures across a wide range of stable weather stations in the Northern and Southern Hemispheres over a century (1917-2018). In particular, we test whether CO2 emissions/concentrations Granger-cause the underlying factors of the di erent quantiles of the distribution of changes in temperature, and find that they exhibit much higher predictivepower on large negative and medium (lower and middle quantiles) than on large positive changes (upper quantiles). These findings are novel in this literature and complement recent results by Gadea and Gonzalo (2020) who document the existence of steeper trends in lower temperature levels than in other parts of the distribution.engAtribución-NoComercial-SinDerivadas 3.0 EspañaGlobal WarmingCo2 EmissionsQuantile Factor ModelsGranger Causalityworking paperC31C33Q54EconomíaDT/0000002018