Research Seminars @ Illinois

Abstract: We develop asymptotic F and t tests for nonlinear cointegration regressions, where the regressors are endogenous and asymptotically homogeneous transformations of I(1) processes. These transformations encompass a broad class of functions, including distribution-like functions, logarithmic functions, and polynomial-type functions. The proposed transformed-augmented OLS (TA-OLS) framework enables more flexible and parsimonious model specifications than existing “fixed-b” methods. The theory also accommodates nonstationary, fractionally integrated drivers for the regressors. Monte Carlo simulations confirm the finite-sample reliability of the proposed tests. In an empirical application to the carbon Kuznets curve, the TA-OLS approach provides new insights into the relationship between environmental pollution and economic activity.