Evaluations and Adjustments

At the end of our last class, we obtained the one-period forecasts for Motorland store. Alte raises a question, “In Chapter 2, a long series of data is analyzed only to obtain one forecast value for the 13th period. All other ‘forecasts’ are current values that are known to us. Why do we need a long time series then?”

Dr. Theo commends Alte on asking a good question and explains, “We need a long series because we have to compare the forecast values with the actual values in order to evaluate the accuracy of our forecast models.” Mo exclaims, “Oh yes, my boss did not want to under-order or over-order by more than 20 percent, so we had better check on our model.” Dr. Theo says, “Right, so the error interval ...

Get Seeing the Future now with O’Reilly online learning.

O’Reilly members experience live online training, plus books, videos, and digital content from 200+ publishers.