The case for trend‐stationarity is stronger than we thought

In DeJong and Whiteman (1991a), the authors concluded that 11 of the 14 macroeconomic time-series originally studied by Nelson and Plosser (1982) supported trend-stationarity. Phillips (1991) criticizes this inference, claiming that their procedure is biased against integration, and that their results are sensitive to model and prior specification. However, Phillips' alternative models and priors bias his results in favor of integration; despite these biases, Phillips' own findings indicate that the data provide the greatest relative support to trend-stationarity. This result is similar to their own (1989, 1990, 1991b) findings concerning the sensitivity of their results; the trend-stationarity inference is remarkably robust. Copyright 1991 by John Wiley & Sons, Ltd.