A0257
Title: A negative moment bound for integrated autoregressions with polynomial time trend and its applications
Authors: Shu-Hui Yu - Institute of Statistics (Taiwan) [presenting]
Abstract: A moment bound is established for the inverse of the normalized Fisher information matrix in an integrated autoregressive model with polynomial time trends. This bound serves as the primary technical tool to derive an asymptotic expression for the mean squared prediction error (MSPE) of the least squares predictor. The derived expression is noteworthy because it provides the first precise assessment of nonstationarity, model complexity, and model over-specification impacts on the MSPE, forming a solid theoretical foundation for certain model selection criteria.
