Unit Root Test (PP) with Stata (Time Series)

The DF test assume that the \({{u}_{t}}\)  are i.i.d.

The ADF test adjust the DF test to take care of possible serial correlation in the \({{u}_{t}}\)   by adding the lagged difference term of the regressand.

Phillip and Perron (1988) use nonparametric statistical method to take care of serial correlation in \({{u}_{t}}\)  without adding lagged difference term.

The pperron uses Newey-West standard errors to account for serial correlation.

To perform the PP test, we will use same data when we perform the ADF test before, Macro_data.

To perform PP test for gdp (use the same lag selection as in ADF test before);

pperron gdp,trend lags(2)reg

The results for PP test show that the \({{t}_{s}}=-2.029\) , and if we choose significant level \(\alpha =0.05\), the \({{t}_{c}}=-3.463\).

The decision is we fail to reject the null hypothesis for unit root. 

That means the series of gdp (in level) is contained unit root processes and thus it’s nonstationary.
 
   

 

Now, perform the PP test for variable gdp  in first difference form;

pperron D.gdp,lags(2)

The results for PP test show that the \({{t}_{s}}=-6.657\) , and if we choose significant level \(\alpha =0.05\), the \({{t}_{c}}=-2.90\).

The decision is we successful reject the null hypothesis for unit root.

That means the series of gdp in first difference is stationary.