Understanding mean error, mean square error, and mean absolute percentage error.
Three of the most common measures of forecast accuracy are: mean error, mean square error, and mean absolute percentage error.
Briefly describe the circumstances when you think each might be the most appropriate measure for selecting the 'best' trend curve.© SolutionLibrary Inc. solutionlibary.com 9836dcf9d7 https://solutionlibrary.com/business/strategy-and-business-analysis/understanding-mean-error-mean-square-error-and-mean-absolute-percentage-error-2tb
... or for testing alternative models to see if you can find a method that eliminates the bias. Unfortunately if there is no bias in the forecast and you get a ME near to 0, you can't tell if it is truly accurate because a series of equally significant + and - values may serve to cancel each other out. The general rule is that if you are happy with a good average forecast and you are not so concerned about individual data points, it's okay to rely on the mean error. A good example would be forecasting sales results of individual salespersons for use in budgeting activities. An upward bias in the forecast would be serious because you would be consistently coming in under plan. Errors in individual results are okay because they tend to cancel ...