Im trying to perform some time series analysis on 2015 and 2016 monthly recorded data to see what method is best for forecasting 2016 monthly values for the remainder of the year. The data has an annual cycle and I already have values for the first 9 months out of 2016. Using Winter's method, I forecasted values for the first 9 months of 2016 and summed up the absolute differences between these forecasted values and the actual 2016 values I already had. From there I calculated the 2016 monthly forecasted value to be off from the 2016 monthly actual value by a 12% average. I then tried a second method where I summed up the absolute differences between the 2015 actual monthly average 2016 actual monthly value. Long story short this elementary method resulted in predictions being off by an average of 6%. My question is how is the Winters error percentage so much higher than using a straghtforward average method. I should also note the MAPE predicted in Minitab using Winters for fitted 2015 monthly data was 2%. Can anyone make sense of all this for me? It would be greatly appreciated.
Hi, if your time series has annual cycle, you can't use Winters method, because you have to wait actual data of all 2016 months.
You could use Brown or Holt method with data available, or I suggest you moving average with 1 or 2 periods.
Anyway, now we are on February 2017, if you still have the actual value of 2015 and 2016 you can try again to use Winters for predict January and then check your accuracy with axtual data of 2017.
In my opinion, as you did, that is match the average of 2015 with data of 2016 is a similar way to use moving average with 12 periods.
Let me know if I helped you and what you think.