What is Forecasting for Classical Data?
Forecasting is a systematic effort to anticipate future events or conditions. The most well known type of forecast may be that of the meteorologist who prepares daily weather forecasts that help us decide how to dress each day and whether to take an umbrella when we leave for work in the morning. Other common forecasts are those that anticipate future economic conditions, traffic patterns, and even the size and number of classrooms that will be needed in local schools.
Forecasting is different from predicting, although both strategies involve statistical projections. In a prediction framework, the results of a statistical analysis are used to make decisions. Under a forecasting framework, statistical projections are seen as merely the beginning of a more involved decision-making process.
Forecasting is also more inclusive than prediction. Prediction models can only account for measurable factors, or variables for which data actually exist. Forecasting models are not limited by data availability. The heart of a forecasting process, in fact, is often the discussion that takes place after statistical projections are complete. These discussions can address a much wider range of factors, including practice and policy concerns for which there may never be objective data.
Time Series Forecasting
Time-series forecasting is a forecasting method that uses a set of historical values to predict an outcome. These historic values, often referred to as a "time series", are spaced equally over time and can represent anything from monthly sales data to daily electricity consumption to hourly call volumes.
Time-series forecasting assumes that a time series is a combination of a pattern and some random error. The goal is to separate the pattern from the error by understanding the pattern's trend, its long-term increase or decrease, and its seasonality, the change caused by seasonal factors such as fluctuations in use and demand.