- This topic has 7 replies, 3 voices, and was last updated 8 years ago by
.
- Posts
I cant figure out where the trend value of 149 comes from for quarter 1 of 2013 using a best fit approach.
“giving a predicted value of 134(149-14.7)”
I understand the 14.7 is a seasonal variation.
Also for quarter 3 of 2013 I do not understand where the trend value of 151 comes from?
You take the last trend value calculated (the last moving average) and add on the required number of awaauva (trend increase per season).
The last trend value is 146.25-14.70= 131.55, and not the 134 given in the solutions? Forgive if this may seem stupid but I am having difficultly understanding this.
As the answer says, there is no set way of extrapolating the trend and it has hypothesized 49
The trend was zero for a while, then increased. We are estimating the future, so it might be fair to look at the increase from 142.25 to 146.25, the latest 4 trend figures (3 ‘jumps’). That is in increase of (146.25 – 142.25)/3 = 1.3
2013 quarter 3 is 3 seasons on from the last season for which there is a trend figure (2012 quarter 4), so add 1.3 x 3 = 4 to 146.25 = 150.25.
I know the examiner took 149; he seems to have invented it a bit. The above method or something similar is fine. There are no hard and fast rules for predicting the future.
I would like to ask how to find the figures in “trend” column, i.e. how to calculate 131.00, 131.75, etc. I checked the answer, it mentions that they are the average of first two quarters, but I cannot figure it out.
Many thanks
The first step would have been to average the first four seasons:
(110 + 160+ 155+ 96)/4 = 130.25.
However, that average would fall between quarter 2 and quarter 3
Next, move down one quarter and average those four:
(160+ 155+ 96+116)/4 = 131.75
This average is between quarters 3 and 4
Average the two averages: (130.25 + 131.75)/2 = 131
This would line up with Quarter 3
- The topic ‘Dec 2013 Q 2 a- Time Series Analysis’ is closed to new replies.