Stationarity Tests

Stationarity tests are used to determine if a time series has constant statistical properties over time, such as a constant mean and variance. Many quantitative trading models assume stationarity, but financial time series like crypto prices are often non-stationary.

If a series is non-stationary, models may produce misleading results, leading to poor trading decisions. Common tests include the Augmented Dickey-Fuller test.

Identifying whether a series is stationary or needs transformation, such as differencing, is a critical step in building robust quantitative models and forecasting tools.