# Patterns in Time Series Analysis

A *time series* is just a collection of data on attribute values over time. Time series analysis is performed in order to predict future instances of the measure based on the past observational data. If you want to forecast or predict future values of the data in your dataset, use time series techniques.

Time series exhibit specific patterns. Take a look at the following figure to get a better understanding of what these patterns are all about. *Constant* time series remain at roughly the same level over time, but are subject to some random error. In contrast, *trended* series show a stable linear movement up or down.

Whether constant or trended, time series may also sometimes exhibit *seasonality* — predictable, cyclical fluctuations that reoccur seasonally throughout a year. As an example of seasonal time series, consider how many businesses show increased sales during the holiday season.

If you’re including seasonality in your model, incorporate it in the quarter, month, or even six-month period — wherever it’s appropriate. Time series may show *non**–**stationary processes* — or, unpredictable cyclical behavior that’s not related to seasonality as a result of economic or industry-wide conditions. Since it’s not predictable, non-stationary processes can’t be forecasted. You must transform non-stationary data to stationary data before moving forward with an evaluation.

Take a look at the solid lines. These represent the mathematical models used to forecast points in the time series. The mathematical models represent very good, precise forecasts because they’re a very close fit to the actual data. The actual data contains some random error, thus making it impossible to forecast perfectly.

The following figure shows you the different time series types, and the mathematical models that describe them.