Data considerations for Augmented Dickey-Fuller Test

To ensure that your results are valid, consider the following guidelines when you collect data, perform the analysis, and interpret your results.

The series length must be 20 or more

The minimum sample size in simulations to develop accurate estimates of the distribution of the test statistics was 20. For more information, see MacKinnon (2010)¹.

Because you usually use the augmented Dickey-Fuller test to find the order of differencing for an ARIMA model, also consider the information in the following topic: Data considerations for ARIMA.