Forecast based on last month



I'm working on ETL created by someone else connecting 9 data sets.

This is for a forecast, the March 29th, 30th and 31st are missing as the Feb only have 28 days.

I'm able to join the Domo calendar and get the missing dates.

For query we are using the date operator and adding the next month to it. I'm not sure how to

forecast for March 29th, 30th and 31st.

Any Suggestions?


  • GrantSmith

    It depends on how you want to do the comparison. Typically when you look at period over period Domo will just subtract a 1 month interval which would make the 29th, 30th and 31st compare to Feb 28th. There are other options where you'd look at 28 days to keep the days of the week aligned (comparing the same day of the week together) but you wouldn't be comparing the same day number to each other and could likely be comparing the same month to itself (29th, 30th and 31st).

    You could create a dataset which would subtract a month (instead of adding a month) but then use the original date as your projected date and the subtracted date as your current date.

    Your business use case would decide how you want to your forecasting model to be designed.

    **Was this post helpful? Click Agree or Like below**
    **Did this solve your problem? Accept it as a solution!**
  • Hiraayub

    I will try this and see if it's work.

    Thank you!