Interpretation ADF-test results
Dear all,
I could use some help with the correct interpretation of these ADF-test results on stationarity. Based on my time series I go results indicating 3x FALSE and 1x TRUE as well as 2x FALSE and 2x TRUE, see also the screenshot below. This leaves me wondering what prevails. Can I conclude that the time series are non-stationary or is the opposite the case. Please advice:).
Comments
Based on the output table, the time series exhibits a deterministic trend over time, so it is stationary with a deterministic trend.
If you were to detrend the time series, the detrended time series is stationary.
Thank you for the input Mohamad. However, these results were based on returns instead of log returns. In case of log returns I got slightly different outcomes, please see below. Is it still right to conclude that the time series are 'stationary with a deterministic trend and that no unit root is present'? Please advice.
In the table above, the series is stationary even without a trend or constant, so you may conclude it is stationary (period). Whether log returns has a deterministic trend or not, it does not, as the 1st restrictive scenario (no constant) exhibits stationarity.
Thank you
Please sign in to leave a comment.