# time series

This topic contains 3 replies, has 2 voices, and was last updated by  John Moffat 1 week, 6 days ago. This post has been viewed 106 times

Viewing 4 posts - 1 through 4 (of 4 total)
• Author
Posts

• fauz
Participant
• Topics: 2
• Replies: 2

The following variations and trend have been calculated for sales over a period of time using additive model:

seasonal variation
quarter 1 +25
quarter 2 -10
quarter 3 -30
quarter 4 +15

trend +50 per quarter
the last known trend reading was taken in year 3,quarter 3 and was \$1750

Question:what would be the time series value for year 4 quarter 3?

John Moffat
Keymaster
• Topics: 3
• Replies: 5796

The last trend reading was year 3 quarter 4. So, by year 4 quarter 3, three more quarters have passed. So the trend forecast will be 1750 + (3 x 50) = 1900.

Quarter 3 has a seasonal variation of -30, so the actual forecast for year 4, quarter 3 will be 1900 – 30 = 1870.

fauz
Participant
• Topics: 2
• Replies: 2

John Moffat
Keymaster
• Topics: 3
• Replies: 5796

Sorry – my fault.

There are four more quarters (not three) from year 3 quarter 3 to year 4 quarter 3.

So the trend forecast will be 1750 + (4 x 50) = 1950.

Quarter 3 has a seasonal variation of -30, so the actual forecast is 1950 – 30 = 1920.

Viewing 4 posts - 1 through 4 (of 4 total)

You must be logged in to reply to this topic.