# Stationarity Most time-series models assume that the underlying time-series data is **stationary**. This assumption gives us some nice statistical properties that allows us to use various models for forecasting. **Stationarity** is a statistical assumption that a time-series has: * **constant mean** * **constant variance** * **autocovariance does not depend on time** More simply put, if we are using past data to predict future data, we should assume that the data will follow the same general trends and patterns as in the past. The general statement holds for most training data and modeling tasks. #### Constant Mean  #### Constant Variance  #### Autocovariance does not depend on time  Sometimes we need to transform the data in order to make it stationary. However, this transformation then calls into question if this data is truly stationary and is suited to be modeled using these techniques. --- Date: 20211230 Links to: [Time-Series-MOC](Time-Series-MOC.md) Tags: References: * [Great set of visuals](https://www.analyticsvidhya.com/blog/2015/12/complete-tutorial-time-series-modeling/)