Announcement

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

  • Analytical derivative of a function

    Dear all,

    I would like to ask if there is a command for getting analytical derivatives of a function in Mata? Perhaps something similar to the diff() function of the Symbolic Math Toolbox in Matlab.

    I would like to use such a command for Maximum Likelihood Estimation with moptimize().

    Many thanks and kind regards,

    Alexander Mosthaf

  • #2
    I could definitely be wrong, but I'm fairly sure Mata does not have such a function.

    Comment


    • #3
      Originally posted by Jesse Wursten View Post
      I could definitely be wrong, but I'm fairly sure Mata does not have such a function.
      Browsing the mata manual gave me the same feeling, only numerical derivatives are discussed (with the deriv()command), but not symbolic computation.

      Ps : I personnaly use SageMathCloud to do symbolic computation, but that's just my 2cents..

      Comment


      • #4
        Many thanks for your answers! (Obviously I could also not find a possibility to get analytical derivatives in Mata.) Some time ago, I heard that such a function would be available in future versions of Stata/Mata. So I just wanted to make sure that I have not missed anything.

        Comment


        • #5
          Alexander,
          There is a thread in the General section named "whishlist for Stata 15", I guess you could ask there whether it is programmed for a close future. Statacorp employees often answer on that thread.

          Comment


          • #6
            I don't see any StataCorp employees in that thread. That is not surprising as StataCorp has a very strict policy of not saying what will be in future versions or when these version will appear. That does not mean that they don't read that thread. I suspect they do so with interest, but they won't comment on it.
            ---------------------------------
            Maarten L. Buis
            University of Konstanz
            Department of history and sociology
            box 40
            78457 Konstanz
            Germany
            http://www.maartenbuis.nl
            ---------------------------------

            Comment


            • #7
              Oh, sorry you're right Maarten, I probably have mistaken with an announcement post, where they detail novelties.

              So bad news for you Alexander, you can't have any clues on whether they will include symbolic calculation soon enough.

              Best,
              Charlie

              Comment

              Working...
              X