Contradicting Stationarity results (ACF and Unit Root Tests)

[Guest]

I have done the Phillips-Perron Test and Augmented DF Test on my data and both point out that I have a stationary time series data. However, my ACF says otherwise. A symptom of non-stationarity is that the plot of the estimated autocorrelations dies down very slowly with increasing j, this provides a visual symptom of nonstationarity. How should I decide if my data are stationary? Below is my ACF:

My unit root tests results are as follows, and since I can't determine if there's a trend visually, I tried both specification with and without trend:

. pperron csad

Phillips-Perron test for unit root Number of obs = 2347
Newey-West lags = 8

---------- Interpolated Dickey-Fuller ---------
Test 1% Critical 5% Critical 10% Critical
Statistic Value Value Value
------------------------------------------------------------------------------
Z(rho) -1084.810 -20.700 -14.100 -11.300
Z(t) -25.541 -3.430 -2.860 -2.570
------------------------------------------------------------------------------
MacKinnon approximate p-value for Z(t) = 0.0000

. pperron csad, trend

Phillips-Perron test for unit root Number of obs = 2347
Newey-West lags = 8

---------- Interpolated Dickey-Fuller ---------
Test 1% Critical 5% Critical 10% Critical
Statistic Value Value Value
------------------------------------------------------------------------------
Z(rho) -1103.166 -29.500 -21.800 -18.300
Z(t) -25.787 -3.960 -3.410 -3.120
------------------------------------------------------------------------------
MacKinnon approximate p-value for Z(t) = 0.0000

. dfuller csad, lag(0)

Dickey-Fuller test for unit root Number of obs = 2347

---------- Interpolated Dickey-Fuller ---------
Test 1% Critical 5% Critical 10% Critical
Statistic Value Value Value
------------------------------------------------------------------------------
Z(t) -23.996 -3.430 -2.860 -2.570
------------------------------------------------------------------------------
MacKinnon approximate p-value for Z(t) = 0.0000

.

[Stefano Grillini]

You data are stationary according to the ADF test, since the statistic is smaller than your three critical values. See other reply for clarification

[Guest]

Thank you for your help, Stefano!