Time Series Forecasting - For Aggregated measure and sub variables

#1

Background:

There is an aggregated measure represented by a variable A, modeled as a time series from a process. There was a need forecast A and also to find out the historical amount of data of A that is the best reflector of future values of A (as there was a data storage capacity issue). Using a combination of sliding window regression technique and ARIMA, it is found that the size of the sliding window out of different window sizes tried, is 100 (as it gave lesser MAPE than the rest of the ones). So the past 100 values of A is a better reflector of future. This forecast was successful.

“A” aggregation comes from B and C of the same process, such that A=B+C and there is a need to predict these variables as a percentage of A. B and C are modeled as a time series.

Questions:

Can the same window size (100 as determined in the previous step) be used to predict B and C as a percentage of A using ARIMA?

Since B and C are expressed as a percentage of A, is it ok to predict B and then calculate C as (100-B) as time series B and C will be mirrored, meaning the same ARIMA parameters will hold good?

Any thoughts and recommendations?

Note: Forecasting B and C and then calculating A may be a good option, but following a deductive approach is important here

#2

Forecast individual variables and deduce the variable A using the forecasted value. This is ideally a better approach. But depends as well. For example we are trying to predict A which is nothing but B/C, If i have information of B and C forecasting A as B/C vs forecasting B and C and then calculating B/C- i would choose the second approach. I am assuming i understood your question. If not apologize.

#3

Hello,

Thank you for your response. Forecasting B and C and then adding it up makes a computation much more complex and time-consuming.

Forecasting A should be done on multiple objects and then choose the object which has maximum forecasted A and then get its forecasted B and C. This is the problem statement. May be you could share your insights now.?