Working Paper: NBER ID: w15292
Authors: Ulrich Müller; Mark W. Watson
Abstract: Standard inference in cointegrating models is fragile because it relies on an assumption of an I(1) model for the common stochastic trends, which may not accurately describe the data's persistence. This paper discusses efficient low-frequency inference about cointegrating vectors that is robust to this potential misspecification. A simple test motivated by the analysis in Wright (2000) is developed and shown to be approximately optimal in the case of a single cointegrating vector.
Keywords: cointegration; low-frequency inference; stochastic trends
JEL Codes: C32; E32
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Common stochastic trends (C22) | Persistence of cointegrating vectors (C32) |
| Low-frequency inference methods (C51) | Robustness against fragility of standard inference methods for cointegration (C22) |
| Low-frequency test derived from Wright's method (C29) | Rejection probability does not depend on the nature of the stochastic trend (C22) |
| Focusing on a single cointegrating vector (C20) | Near-optimal power (H21) |
| Low-frequency version of the test (C22) | Achieves upper power bound when examining the value of the cointegrating vectors (C51) |