0
$\begingroup$

using simple R lines, I have plots of decomposition of a time series.

DCB = decompose(ts_series, "additive")
plot(as.ts(DCB$seasonal))
plot(as.ts(DCB$trend))
plot(as.ts(DCB$random))
plot(DCB)

Not obvious if it's additive or multiplicative so I did both. the outputs are similar.

The "observed", which is the plot of the numbers itself, shows a relatively flat movement. However the trend decomposed, shows an obvious downward trend.

Why are they so different (the Observed vs the Trend)? and the Trend is missing it's beginning and ending parts? Thank you.

$\endgroup$
  • $\begingroup$ I can't read the over-lapping numbers on the y-axes ... $\endgroup$ – Roland Nov 26 '18 at 9:49
  • $\begingroup$ @Roland, the Observed starts from 80,000. $\endgroup$ – Mark K Nov 26 '18 at 9:51
  • 1
    $\begingroup$ I was trying to hint that you should compare the scales of trend and observed. The trend might very well be actually quite flat compared to the season effect, but I can't see the range of the scales in your plots. $\endgroup$ – Roland Nov 26 '18 at 9:53
  • $\begingroup$ @Roland, thank you very much for the hint! brilliant point! more hints, please? :) $\endgroup$ – Mark K Nov 26 '18 at 20:45
  • $\begingroup$ @Roland, would you consider to turn the comment as the answer, so that we can close this question? thank you. $\endgroup$ – Mark K Nov 29 '18 at 1:05

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.