0

I am modeling a forecast for product categories using auto Arima in R. I'm getting MAPE of between 9-15% on average. We don't have any historical records of forecasts vs actual so I don't know how this racks up in terms of effectiveness. Any thoughts on what a good error target is?

André.B
  • 1,290
  • 6
  • 20
MrMooseIsSick
  • 1
  • 3
    If by error target you mean prediction error, then it depends entirely on what is acceptable error for the system you are working in. What do you mean forecasts v.s. actual? – André.B Sep 30 '19 at 23:04
  • 2
    Possible duplicate of [Is my model any good, based on the diagnostic metric ($R^2$/ AUC/ accuracy/ RMSE etc.) value?](https://stats.stackexchange.com/questions/414349/is-my-model-any-good-based-on-the-diagnostic-metric-r2-auc-accuracy-rmse) – Sycorax Oct 01 '19 at 01:39

0 Answers0