How to interpret Panel Unit Root test result?

[irisbear]

Hi there, I am a new user for Eviews. Appreciate if anyone could help out on interpreting the below results for 121 time series by using panel Unit Root test. Is the result random walk or mean reverting? Thanks.


Exogenous variables: Individual effects
Automatic selection of maximum lags
Automatic lag length selection based on SIC: 0 to 21
Newey-West automatic bandwidth selection and Bartlett kernel

Cross-
Method Statistic Prob.** sections Obs
Null: Unit root (assumes common unit root process)
Levin, Lin & Chu t* 2.81302 0.9975 121 466273

Null: Unit root (assumes individual unit root process)
Im, Pesaran and Shin W-stat -0.86649 0.1931 121 466273
ADF - Fisher Chi-square 339.874 0 121 466273
PP - Fisher Chi-square 330.316 0.0001 121 470085

** Probabilities for Fisher tests are computed using an asymptotic Chi
-square distribution. All other tests assume asymptotic normality.

[irisbear]

Hi, in addition, I am using Eviews 9S.

To be more specific to my question, Levin is testing common unit root process which is similar to an ADF test on multiple time series. For ADF test on single time series, we could compare the t-statistic against the Model 0,1,2 critical value to reject or accept the null hypothesis.

My question is are we comparing the same table against Levin, Lin & Chu result? Per above, the statistic is 2.81302 which is higher than ADF table. It means we should reject the null hypothesis? Hence the stock prices are mean reverting?

Appreciate if someone could guide me to the right direction as I am stuck in the middle of my thesis data analysis process.

Many thanks!

[NicolasR]

Hi,

The LLC test estimates a single coefficient of the lagged variables to perform the unit root test while the IPS test estimates one coefficient of the lagged variables for each series and then calculates the test with those. For what i remember the LLC has a diferent distribution than the ADF.

Regards,