0

I am using FBProphet to predict future values in my monthly datasets.

I noted the following in the documentation.

Note that with weekly or monthly aggregated data, many holiday effects will be well-captured by the yearly seasonality, so added holidays may only be necessary for holidays that occur in different weeks throughout the time series.

Questions

  1. For monthly data, does that mean I do not need to add holidays that fall on a fixed date? For example, Christmas day.

  2. For Chinese New Year, it falls on either Jan or Feb each year. So do I assign 1 Jan 20XX or 1 Feb 20XX to Chinese New Year, depending on the year?

Thanks for your advice.

nuttychan
  • 21
  • 4
  • 1) Yes. 2) Right. You might not have to include Chinese New Year, either, unless you expect the month that Chinese New Year falls in to affect the variable you wish to forecast. You know your data better than we do. Is there a good reason to think it's important? – Nick ODell Nov 24 '22 at 03:24

0 Answers0