2019/08/07 by Yicong Lin, Lin, Yicong, Hanno Reuvers +1
Mathematics · #Advanced Statistical Methods and Models #Econometrics (econ.EM) #FOS: Computer and information sciences #FOS: Economics and business #Methodology (stat.ME) #Statistical Methods and Inference
paper · pdf · doi:10.48550/arxiv.1908.02552
openalex publication_date 2019/08/07 · openalex created_date 2021/02/01 · openalex updated_date 2026/07/28
This paper develops the asymptotic theory of a Fully Modified Generalized\nLeast Squares estimator for multivariate cointegrating polynomial regressions.\nSuch regressions allow for deterministic trends, stochastic trends and integer\npowers of stochastic trends to enter the cointegrating relations. Our fully\nmodified estimator incorporates: (1) the direct estimation of the inverse\nautocovariance matrix of the multidimensional errors, and (2) second order bias\ncorrections. The resulting estimator has the intuitive interpretation of\napplying a weighted least squares objective function to filtered data series.\nMoreover, the required second order bias corrections are convenient byproducts\nof our approach and lead to standard asymptotic inference. We also study\nseveral multivariate KPSS-type of tests for the null of cointegration. A\ncomprehensive simulation study shows good performance of the FM-GLS estimator\nand the related tests. As a practical illustration, we reinvestigate the\nEnvironmental Kuznets Curve (EKC) hypothesis for six early industrialized\ncountries as in Wagner et al. (2020).\n