How to Get Good Forecasts from Bad Data

Ellen’s article has three key points. She advises forecasters to accept the idea that data do not have to be perfect. Instead of changing unreliable data used for reporting, she suggests that forecasters create a second set of data to be used for forecasting. She also makes the point that a company’s fiscal calendar, product groupings, and location hierarchies may not be a sound basis for forecasting. Instead of relying on these, she establishes her own calendars, product groups, and location hierarchies with the specific forecasting task in mind. Finally, she observes that executives want forecasting problems to go away. But if they determine what actual problems need to be solved, there might be a way to use forecasting as part of the solution. Copyright International Institute of Forecasters, 2007