Following are the results from Fisher-type unit-root test for RDI (dependent variable). How do you interpret it?

Based on Phillips-Perron tests:

Ho: All panels contain unit roots           Number of panels       =    100
Ha: At least one panel is stationary        Avg. number of periods =  10.74

AR parameter:    Panel-specific             Asymptotics: T -> Infinity
Panel means:     Included
Time trend:      Included
Newey-West lags: 1 lag

                                     Statistic               p-value

Inverse chi-squared(196)     P       207.1519                 0.2788
Inverse normal               Z        2.0005                  0.9773
Inverse logit t(389)         L*        0.5211                0.6987
Modified inv. chi-squared    Pm        0.5633                 0.2866

P statistic requires number of panels to be finite.
Other statistics are suitable for finite or infinite number of panels.

1 Answer 1


So in the upper left you see which hypothesis you are testing. In this case your null is that all of your panels contain a unit root. In the lower half of your output you see the outcome of the test statistics (P,Z,L*,Pm) to test this hypothesis and their associated p-value. All of the p-values are relatively large. If we were to use a 10% level of statistical significance (5% is more commonly used), you can see that all of your p-values exceed this threshold. So on the basis of this we cannot reject the null hypothesis. Which means that your panel data contains a unit root.


Not the answer you're looking for? Browse other questions tagged or ask your own question.