Components of a Time Series:

The Components of a Time Series

• For long-term forecasts (2 or more years) seasonal and random components are usually avoided, and major focus is put on trend component with minor emphasis on the business cycle component.
• For medium-term forecasts (1 to 24 months), the trend factor becomes less important, and focus is mostly on seasonal and business cycle factors with minor emphasis on the random component.
• For short-term forecasts (1 to 5 weeks), the focus is on random fluctuations and any short-duration regular movements (weekly or daily patterns).