Announcement

Collapse
No announcement yet.
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • Is it true that there is no built-in stata program for seasonal adjustment?

    Can't believe that! Hope I'm wrong. All I found is a stata journal article "Menu-driven X-12-ARIMA seasonal adjustment in Stata" of Q. Wang & N. Wu (2012). But the links in this article are outdated. Has anybody relevant information about this topic?
    Thanks a lot!
    Rene

  • #2
    it is not entirely clear what you are looking for; before clarifying, you should start with
    Code:
    hsearch seasonal

    Comment


    • #3
      Sorry, "hsearch seasonal" did not help to solve my problem. I have a monthly time series and want to eliminate the seasonal figure from this time series. There are various methods. I could tinker a procedure, but would first like to know, whether there is stata code for purging my data from seasonality...

      Comment


      • #4
        You may wish to type:
        Code:
        . help tssmooth shwinters
        and check whether the Holt-Winters with a seasonal component may do the trick to you.

        There are 2 versions (multiplicative, the default; and additive).

        Best regards,

        Marcos

        Comment


        • #5
          Thanks a lot Marocs! "tssmooth shwinters ln_CPI_sa = ln_CPI, period(12)" is doing the job well!
          Best wishes,
          Rene

          Comment


          • #6
            Rene Macon Thank you for sharing here the way you used - tssmooth shwinters - to tackle the problem.
            Best regards,

            Marcos

            Comment


            • #7
              Marcos Almeida please can you help me?
              I did everything according the manual (first command tsset, after tssmooth shwinters) but Stata told me that my data are not feasible.

              My time series data are 13 years long and they are on quarter frequency (52 observations in total).

              Do you have an idea where the problem is?

              Thank you for your answer Pavla

              Comment


              • #8
                Perhaps you should start a new thread. In any case, please make sure to inform what you typed as well as what Stata gave you back.
                Best regards,

                Marcos

                Comment

                Working...
                X