SUMMARY
Trend differences
A trend is an increase or decrease in a series that persist for an extended time.
Trends can be difficult to detect when a series has significant randomness and seasonality.
The first differences are:
Yt - Yt-1 = Change from period t-1 to t
Forecast is then:
Yt = Yt-1 + mean of the differences
Demand of serie is in the Examples Data as Serie C (Strong Trend).
We calculate the mean of the differences:
It is the sum of the differences divided by the number of entries
(70-6+58+ ... +54+82 ) / 47 = 698 ~= 14.85
Forecast is then:
Yt = Yt-1 + mean of the differences
Y49 = Y48 + 14.85
Y49 = 1166 + 14.85 = 1180.85 ~ 1181
BETA
All rights reserved 2025. ©futurera