I know it might be trivial but does the density of daily values impact the forecast accuracy? For example, if a call center receives less than 50 calls for weekdays and less than 10 calls for weekend, is the forecast accuracy diminished compared with the case that the call center receives more than 100 calls during weekdays and more than 50 calls during weekend. I saw this trend in a few data sets but I wanted to make sure that daily density impacts forecast accuracy.

  • $\begingroup$ You might be confusing accuracy with precision. A forecast can be perfectly accurate (meaning that in the long run its average value will equal the actual average) but, due to small numbers of weekend calls, will be imprecise (meaning the outcomes will vary quite a bit above and below the forecasts) only because the actual outcomes on the weekends will be highly variable due to chance alone. $\endgroup$ – whuber Oct 22 '14 at 15:08
  • $\begingroup$ @whuber Thank you for your note. To be more accurate, I always find forecast accuracy for weekdays separately from weekend. This means that I get an average for weekdays and and average for weekends. My question still remains that if lower density for either weekday or weekend impacts forecast's accuracy (in your language I believe it is imprecise) $\endgroup$ – user12 Oct 22 '14 at 15:33
  • $\begingroup$ You would do well to adopt my language because it is the one everyone uses. If you continue to use "accuracy" where "precision" is meant, you risk obtaining truly confusing answers. $\endgroup$ – whuber Oct 22 '14 at 15:40
  • $\begingroup$ @whuber Thank you for correcting me. could you please give me the right definition for "precision" and "accuracy" so I dont use them wrong? Thanks so much $\endgroup$ – user12 Oct 22 '14 at 15:44

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.