Forecasting

Time series analysis relies on historical data and attempts to project historical patterns into the future, assuming that the same patterns will continu. The patterns that may appear in time-series data are trend, cyclical, seasonal, and random variations. Trend is a long-term growth or decay pattern. Cyclical pattern is a long-term variation around the trend. Seasonal pattern refers to short-term variation around the trend. Random refers to irregular variation (Management science information systems, Mohsen Attaran).

The models include all the well-known techniques from simple smoothing, Holt trending and Holt-Winters seasonal models. The algorithms and model interpretations are documented in the book entitled 'Forecasting Systems for operation management' by Delurgio and Bhame.You can try a free version. It has all the features of forecasting.