Test Statistics for Nonparametric Cointegrating Regression Functions through Subsampling

Nonparametric cointegrating regression models have been extensively used in financial markets, stock prices, heavy traffic, climate datasets, and energy markets. Models with parametric regression functions can be more appealing in practice compared to non-parametric forms but do result in potential functional misspecification. Despite the rich literature on developing a model specification test for parametric forms of regression functions, proposing a suitable test statistic under endogeneity and long or semi-long memory structure has been complicated. In this talk, I introduce multiple test statistics for this framework and approximate their sampling distributions by the subsampling methods. I demonstrate the properties of test statistics and explain how they can address important practical problems such as regressing CO2 on GDP of countries related to Carbon Kuznets Curves.