作者: Ismael Sánchez
DOI: 10.1007/BF02777226
关键词:
摘要: This paper compares the performance of classical and recent unit root tests based on different estimation procedures, including fitting ARMA models unknown orders. The article also introduces an estimator spectral density function that is model with data previously detrended by GLS. Monte Carlo experiment shows improve their if estimated, instead autoregressive approximation. best results are obtained function.