iSixSigma

trend analysis / time series – help needed!

Six Sigma – iSixSigma › Forums › Old Forums › General › trend analysis / time series – help needed!

Viewing 6 posts - 1 through 6 (of 6 total)
  • Author
    Posts
  • July 24, 2001 at 4:00 am #67665

    denton
    ★ 20 Years ★
    Participant

    I suggest you start with a simple I&MR chart, to see if the process is stable and predictable.  If you see a trend, do Double Exponential Smoothing, store the residuals, and request 3 periods of forecast (about all this technique is good for).  Your residuals are your detrended data.  Do an I&MR on the residuals to see if the data are stable and predictable with the trend removed.

    0
    July 24, 2001 at 4:00 am #67666

    Anonymous
    ★ 10 Years ★
    Participant

    Can not tell much from this alone.
    Is everything else constant e.g. number of staff?
    Stratify the data into sub categories – is one sub category resposible for (most of) the change.
    Are there any special causes for hig /low months, e.g end of budget year, holidays, etc.

    0
    July 24, 2001 at 4:00 am #67668

    Neil Polhemus
    ★ 20 Years ★
    Participant

    I ran it through the Automatic Forecasting procedure in STATGRAPHICS. On the limited amount of data you provided, the winning method was a simple random walk: the forecast for next month equals this month’s amounts. If you provide more history, it is possible that we could do better.

    0
    July 24, 2001 at 4:00 am #27556

    sreedher
    ★ 20 Years ★
    Member

    hello all,
    I am enclosing a data related to my overhead expenses of one of the categories. I am interested to know the results. I am using minitab r-13 for my statistical applications. Month____Bill amtJune 29993
    July 33947
    Aug 29321
    Sep 27775
    Oct 37325
    Nov 36813
    Dec 40267
    Jan 50108
    Feb 49072
    Mar 36275
    Apr 38934
    May 45224More than anything, if anybody can sugest in simple terms, what kind of analysis to be used and when to use, that would help me to great extent.Regards, and thanks in advance
    Sreedher

    0
    July 25, 2001 at 4:00 am #67681

    sreedher
    ★ 20 Years ★
    Member

    yeah I can do that.. but that is quite big and huge cannot send it in this website. If you can aid me with your mailing address, I would post them to you.
    regards
    Sreedher

    0
    July 26, 2001 at 4:00 am #67712

    Neil Polhemus
    ★ 20 Years ★
    Participant

    You can send it to me at P.O. Box 1124, Englewood Cliffs, N.J. 07632, or e-mail a file to [email protected]

    0
  • Author
    Posts
Viewing 6 posts - 1 through 6 (of 6 total)

The forum ‘General’ is closed to new topics and replies.