# Calculating future data based previous data

The sales volume of the next month is predicted by the data in the past. The sales volume is changed greatly from month to month, but the annual fluctuation pattern is almost the same every year. Which of the following is the most appropriate formula that can be used for calculating the sales volume of the next month? Here, Pt+1 is the sales volume predicted for the next month, St is the sales volume of the current month t, and the data is retained for three years.

a) Pt+1 = (St+ St–1+ St–2) / 3
b) Pt+1= St xSt / St–1
c) Pt+1= (St+ St–12 + St–24) / 3
d) Pt+1= (St–11+ St–23+ St–35) / 3

I would love to know the right method to solve this one, some explanations will be greatly appreciated!

• I think it's impossible to decide which model is the best one, on the basis of the information that you give. It may be necessary to look at the annual fluctuation pattern. In general, what you have here is a time series, and there are many good models for time series, depending on the type of series. Oct 14, 2014 at 9:54