Composite Forecast is a combination of blended forecasting methods (such as times series, casual, and/or judgmental) for a particular brand, product category or product, from different forecast streams.
Forecast carried out following composite forecasting technique is based on the same historical data, but uses a different method.
You cant always rely on one forecasting method. Because as the time and situation progress the need or demand for product/services also varies. Hence, many companies will opt to combine the forecasts they practice into a composite forecast.
The objective is to get the most out of each method and identify which alternative was best and then create a single individual forecast. Either you can average the forecasts giving each one equal ranking, or you can rank each one differently, or you can vary the ranking of each forecast over time.
The most popular form of composite forecasting is when a group of judges who give scores for athletes at sporting events like boxing. By averaging the scores of any three judges at a given boxing event, the obtained combined scores produce a more fair chance of being correct as to who the actual winner should be.