Dear all,
I would like to build a model that could forecast the result at the end of a period (e.g. total sales in June) using intra-period and previous-periods results (e.g. sales on 31/05, sales on 1/06, sales on 2/06, etc.) so that I could predict on each day (with some margin of error) what is the expected result at the end of the period given the previous results and, as the time approaches to the end of the period, the margin of error of the forecast would decrease.
Is there anyone who could help me with the selection of the right model? I thought about an n-step ahead model forecast using a value of n that decreases as the end of the period approaches, but I don't know if such model exists and if my idea makes any sense.
Thank you in advance.
I would like to build a model that could forecast the result at the end of a period (e.g. total sales in June) using intra-period and previous-periods results (e.g. sales on 31/05, sales on 1/06, sales on 2/06, etc.) so that I could predict on each day (with some margin of error) what is the expected result at the end of the period given the previous results and, as the time approaches to the end of the period, the margin of error of the forecast would decrease.
Is there anyone who could help me with the selection of the right model? I thought about an n-step ahead model forecast using a value of n that decreases as the end of the period approaches, but I don't know if such model exists and if my idea makes any sense.
Thank you in advance.
