Statistical Stationarity

Statistical stationarity is a property of a time series where its statistical properties, such as mean and variance, remain constant over time. In financial markets, most price series are non-stationary, meaning they exhibit trends and changing volatility levels.

Traders often transform non-stationary data, such as calculating log returns, to make them stationary, which is a requirement for many statistical models and forecasting techniques. If a series is not stationary, models may produce misleading results, as they are not accounting for the changing underlying structure of the data.

Achieving stationarity is a key step in the quantitative analysis of market data. It allows for more robust testing and modeling of market phenomena like mean reversion.