This appendix describes forecasting terminology. It defines the time-series forecasting methods that Predictor uses, as well as other forecasting-related terminology. This appendix also describes the statistics the program generates and the techniques that Predictor uses to do the calculations and select the best-fitting method.
Forecasting refers to the act of predicting the future, usually for planning and managing resources. There are many scientific approaches to forecasting. You can perform “what-if” forecasting by creating a model and simulating outcomes, as with Crystal Ball, or by collecting data over time and analyzing the trends and patterns. Predictor uses the latter concept, analyzing the patterns of a time series to forecast future data.
The scientific approaches to forecasting usually fall into one of several categories:
Time-series — Performs time-series analysis on past patterns of data to forecast results. This works best for stable situations in which conditions are expected to remain the same.
Regression — Forecasts results using past relationships between a variable of interest and several other variables that might influence it. This works best for situations in which you need to identify the different effects of different variables. This category includes multiple linear regression.
Simulation — Randomly generates many scenarios for a model to forecast the possible outcomes. This method works best where you may not have historical data, but you can build the model of the situation to analyze its behavior.
Qualitative — Uses subjective judgment and expert opinion to forecast results. These methods work best for situations for which no historical data or models are available.
Predictor uses time-series and multiple linear regression for forecasting. Oracle Crystal Ball, Fusion Edition uses simulation. Each technique and method has advantages and disadvantages for particular types of data, so often you might forecast the data using several methods and then select the method that yields the best results.
The following section, Classic Time-series Forecasting, describes forecast methods available in Predictor and their uses.
These sections offer information about other data and forecast methodologies: