A1641
Title: On the reliability of bootstrapped cointegration test findings
Authors: Sven Schreiber - Macroeconomic Policy Institute IMK and Free U Berlin (Germany) [presenting]
Abstract: As applied cointegration analysis faces the challenge that (a) potentially relevant variables are unobservable and (b) it is uncertain which covariates are relevant, partial systems are often used and potential (stationary) covariates are ignored. By simulating hypothesized larger systems earlier findings seemed to suggest that a nominally significant cointegration outcome using a bootstrapped rank test in the bivariate sub-system might be due to test size distortions. We review the issue and the claim systematically. We find only mild size distortions, except when the specified data-generating process includes a large borderline-stationary root, reflecting an earlier insight from the literature. It turns out that when revisiting the earlier application of a long-run Phillips curve (inflation and unemployment), the interpretation of the bivariate cointegration test result (for the euro area) hinges on the assumed persistent output gap measure in the background.
