Welcome to the JMulTi Open Discussion

2005-05-09
2014-10-09
  • This is the place to discuss all issues related to the econometric software JMulTi. If there are any bugs discovered they may be brought up in this forum, or, alternatively, they might be put directly in the bug database.
    We hope for a lively feedback!

     
    • Hallo Markus,

      can you tell me please which multivariate modell is estimated in jmulti?Is it VEC or BEKK or etc?
      what is definition of vector stochastic process in these estimation: is it the conditional mean vector plus error vector  or is it only error vector?

      Thank you very much in advance
      Irina

       
    • help for MGARCH is still missing...
      its the BEKK form, the estimation procedure uses the
      error vector which should have mean 0

      m

       
      • Hello Mr. Kraetzig.
        I read that jmulti can estimate parameter in bekk-multivariate garch. I found the code on Java here (mgarchcall.java), but I dont know how to use it. I have several questions :

        1. Where should I put the code?
        2. Can it appear in GUI like univariate garch so I can get the output as shown in help  file? How?
        3. Do I need to install Java 1.5 first?

        Sorry for the "dumb" questions, but I really have no idea about this. I'm very thankful if you would explain this to me.

        Thank you in advance.

        -b-

         
    • LGM
      LGM
      2013-10-21

      Dear Markus,

      I hope this e-mail will find you safe. I was wondering if you know how to solve the invertibility problems (NaN outcome)in testing linearity in STR. I have seen the help menu within JMulti and it says if the elements of the transition variable is close to zero or one then there will be invertibility problems. Does that mean one cannot go further with the procedure of STR using JMulti if there is invertibiility problem? Is there any way of overcoming the problem please? I would be most grateful if you could help on this. Thank you very much for your time and reply. Looking forward to hear from you.

      LGM

       
    • khanhcuong126
      khanhcuong126
      2014-10-09

      Dear Markus Kraetzig,

      Would you pls answer me a foolish question,
      In STR, why I cannot choose both variables x and y at the same time to run linear AR model?
      Thank you very much!

       
  • MagW
    MagW
    2012-03-20

    Hi! Is there any chance to get response how can I estimate MGARCH using JMulti? I saw a few questions, but no answers…
    Thank you :)

    M

     

  • Anonymous
    2012-09-03

    Can I install/use JMulti in Windows 64?

    In SVAR, how can I do bootstrap run faster?

    thank you

    Cláudio