- This topic has 1 reply, 2 voices, and was last updated 12 years ago by
.
- Posts
Question with moving averages was far away my knowledge. I’ve gone through the BPP book and video lectures but still don’t get one thing.
Cany anyone explain me please how the hell I could know that I should reduce the total averages 4.0057 by 0.0016. Did I miss something in books/lectures or it was never said?
the other thing is: average trend? what is that? Again, couldn’t find in any sources available why I should from centre moving average from Q2 2012 subtract moving average from q1 2011 and result divide by 5????
Can anyone help?
The other
The total of the averages should in theory be 4. (for some seasons the variation is more than 1 and for some seasons it is less than 1, but in total it should be 4).
The actual total is 4.0057 (the reason it is not 4 is that all seasons were used to get the moving averages, but the first 2 and last 2 could not be used to calculate seasonal variations). So…… to make it 4, each one has been reduced by a quarter of the different of 0.0057.
Its actually a bit silly, and I am surprised the examiner actually bothered 🙂The trend is the centred moving average, and is what we would expect if there were no seasonal variations. There are many ways you could forecast the trend into the future (you could for example use regression analysis, although it was not expected here – there were not enough marks to make it worthwhile).
What the examiner had done is say that over the whole period, the trend has grown from 1068.75 to 1287.5. So it has grown by 218.75.
This growth has occurred over 5 quarters and so on average it is growing by 218.75 / 5 = 43.75. We can then forecast the trend into the future by assuming it continues to grow by 43.75 per quarter (the figures given are all in thousands, so it is actually 43,750 per quarter).
- You must be logged in to reply to this topic.