New Solution to Time Series Inference in Spurious Regression Problems

摘要: Phillips (1986) provides asymptotic theory for regressions that relate nonstationary time series including those integrated of order 1, I(1). A practical implication related literature on spurious regression is one cannot trust the usual confidence intervals. Therefore it recommended after carrying out unit root tests we work with differenced instead original data in levels. We propose a new alternative using intervals based Maximum Entropy bootstrap explained Vinod and Lopez-de-Lacalle (2009, J Statistical Software). Extensive Monte Carlo simulations show our proposal can provide more reliable conservative than traditional, differencing block (BB) hope to let researchers avoid variables their specifications