Standard unit-root and cointegration tests are sensitive to atypical events such as outliers and structural breaks. Outlier-robust estimation techniques are used to examine the impact of these events on cointegration analysis. The outlier-robust cointegration test provides a new diagnostic tool for signaling when standard cointegration results might be driven by a few aberrant observations. A main feature of the approach is that the proposed robust estimator can be used to compute weights for all observations, which in turn can be used to identify the approximate dates of atypical events. The method is evaluated using simulated data and a Monte Carlo experiment. An empirical example is provided showing the usefulness of the proposed analysis.

Additional Metadata
Keywords Monte Carlo simulation, Robust estimation, cointegration analysis, diagnostics, econometrics, estimating techniques, outliers, structural breaks, unit roots
Persistent URL hdl.handle.net/1765/2146
Series ERIM Top-Core Articles
Journal Journal of Business and Economic Statistics
Citation
Franses, Ph.H.B.F, & Lucas, A. (1998). Outlier detection in cointegration analysis. Journal of Business and Economic Statistics, 459–468. Retrieved from http://hdl.handle.net/1765/2146