Which time series analysis should I use?

You can use Stat > Time Series > ARIMA for any type of time series data. However, Minitab offers alternative analyses that you can use depending on whether your data have a trend or a seasonal component.

Your data do not have a trend or a seasonal component

When your data don't have a trend or a seasonal component, you can use one of the following analyses:
  • Stat > Time Series > Moving Average
  • Stat > Time Series > Single Exp Smoothing

Your data have a trend but do not have a seasonal component

When your data have a trend but do not have a seasonal component, you can use one of the following analyses:
  • Stat > Time Series > Trend Analysis
  • Stat > Time Series > Double Exp Smoothing

Trend Analysis fits a single equation to the data, which works well when the trend follows a consistent shape without shifts or reversals. Double Exponential Smoothing uses a dynamic trend component that works well when the data have cyclical movements, shifts in the trend, or even reversals in the trend.

Your data have a seasonal component

When your data have a seasonal component (with or without a trend), you can use one of the following analyses:
  • Stat > Time Series > Decomposition
  • Stat > Time Series > Winters’ Method

Use Winters' Method when you want to use your time series model to generate forecasts. Usually, you should not use Decomposition to generate forecasts, but it can be useful to examine the components of the time series. For example, you could use Decomposition to communicate time series concepts to management.

By using this site you agree to the use of cookies for analytics and personalized content.  Read our policy