Forecasting - Summary
Forecasting - Summary
There is a set of qualitative approaches to forecasting that includes the jury of executive opinion, the salesforce estimates, the survey of customer intentions, and the Delphi approach. These approaches are fast, inexpensive, flexible, and can integrate large quantities of information. They are extremely sensitive and can be used for difficult forecasting tasks such as long-range forecasting and forecasting turning points. They suffer from the biases, uncertainties, and inconsistencies inherent in the subjective judgments used.
Time-series forecasting, the projection of historical data, is well-suited to short-term forecasts of data containing a clear trend or seasonal or cyclical patterns. It is not capable of forecasting turning points where the
environment changes. One decision in time-series analysis is how much historical data to use and how much reliance to place on older data.
Causal models refer to the introduction of factors that directly or indirectly cause or influence the forecast. It can simply be the identification of leading indicators or the use of a formal regression model. Causal models allow for a deeper understanding of the process. The causal factors can be used to help predict turning points.
There is a set of qualitative approaches to forecasting that includes the jury of executive opinion, the salesforce estimates, the survey of customer intentions, and the Delphi approach. These approaches are fast, inexpensive, flexible, and can integrate large quantities of information. They are extremely sensitive and can be used for difficult forecasting tasks such as long-range forecasting and forecasting turning points. They suffer from the biases, uncertainties, and inconsistencies inherent in the subjective judgments used.
Time-series forecasting, the projection of historical data, is well-suited to short-term forecasts of data containing a clear trend or seasonal or cyclical patterns. It is not capable of forecasting turning points where the
environment changes. One decision in time-series analysis is how much historical data to use and how much reliance to place on older data.
Causal models refer to the introduction of factors that directly or indirectly cause or influence the forecast. It can simply be the identification of leading indicators or the use of a formal regression model. Causal models allow for a deeper understanding of the process. The causal factors can be used to help predict turning points.
Comments
Post a Comment