Panel KSS
Also known as: Panel stationarity test
The Panel KSS test reverses the null hypothesis of unit-root tests: it tests whether variables are stationary (stationarity is the null) versus nonstationary (unit root is the alternative). Introduced by Kwiatkowski et al. (1992) and extended to panels by Hadri (2000), this complementary approach provides robustness when combined with unit-root tests like Panel DF-GLS. Using both tests together reduces the risk of erroneous conclusions about variable persistence.
Key highlights
- Stationarity null provides complementary evidence to unit-root tests
- Higher power in some settings (stationary series with near-unit-root behavior)
- Long-run variance estimation is robust to correlation structure
- Straightforward to implement with standard formulas
Intuition
This section is available to Pro members. Upgrade to Pro
How it works
This section is available to Pro members. Upgrade to Pro
When to use it
Use Panel KSS alongside Panel DF-GLS for robust unit-root testing. When one test rejects its null and the other fails to reject, use both statistics to triangulate evidence. Particularly useful for controversial series (e.g., real exchange rates, unemployment rates) where test results have policy implications.
Strengths & limitations
- Stationarity null provides complementary evidence to unit-root tests
- Higher power in some settings (stationary series with near-unit-root behavior)
- Long-run variance estimation is robust to correlation structure
- Straightforward to implement with standard formulas
- Power depends on bandwidth choice; results sensitive to bandwidth selection
- Cross-sectional dependence can invalidate critical values in panel settings
- Null and alternative are reversed from standard tests; interpretation requires care
- LM statistic one-sided; critical values differ from standard normal distribution
Common pitfalls
This section is available to Pro members. Upgrade to Pro
Applications
This section is available to Pro members. Upgrade to Pro
Frequently asked
What bandwidth should I use in Panel KSS?
Andrews (1991) bandwidth or Newey-West bandwidth are standard. Software typically implements these automatically. Check sensitivity by varying bandwidth parameter.
How do I reconcile conflicting results from DF-GLS and KSS?
If DF-GLS rejects unit root (stationarity) and KSS fails to reject stationarity, both suggest stationarity. If they disagree, examine the time series visually and consider near-unit-root processes. Report both results and discuss potential reasons for disagreement.
Is Panel KSS affected by cross-sectional dependence?
Yes. If units are cross-sectionally dependent, critical values can be substantially different. Use bootstrap or cross-sectionally demeaned versions for robustness.
Can I use Panel KSS with trending data?
Yes. Include trend in the regression specification. Test outcomes under both constant-only and trend specifications for robustness.
Sources
- 1.Kwiatkowski, D., Phillips, P. C., Schmidt, P., & Shin, Y. (1992). Testing the null hypothesis of stationarity against the alternative of a unit root. Journal of Econometrics, 54(1-3), 159-178.
- 2.Hadri, K. (2000). Testing for stationarity in heterogeneous panel data. Econometric Reviews, 19(4), 367-397.
You have read it. What now?
Cite this page
ScholarGate. (2026, June 3). Panel KSS. ScholarGate. https://scholargate.app/econometrics/panel-kss