The analysis of time series is important to try to predict the future.
A time series has usually a tendency and a sazonality.
We can have additive models
Y = T + S + C + e
Or multiplicative models
Y= T * S * C * e
Where T – Tendency, S – Sazonality, C – Cyclical, e – error
The analysis of errors allows to determine which model is a better fit to that particular time series.