Announcement

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

  • DH Granger causality test

    Dear Stata Users,

    In my research thesis, I'am analysing the causal relationships between economic growth and tourism industry applying the Dumistrescu-Hurlin Granger non-causality test through the '.xtgcause' command. Since I have never used Stata before, I need your kind help.

    I use annual logarithmic data (see the attachment) on gross domestic product and gross earnings from accommodation fees (as proxies for economic growth and tourism expansion) in 20 Hungarian counties for the period 2000-2016.

    I would like to conduct the analysis from county to county; however, I have no idea how can I do this in one single step. Should I perform 20 separate analysis using '.if' command, or there is a solution for my issue? I tried to apply the '.bysort' and '.by' commands, but I got an error code r(109).

    Also, before running DH Granger non-causality test, a test for stationarity and unit roots should be applied. May I use '.xtunitroot' command or other way is recommended to get my results?


    Thank you for your kind reply.


    Yours sincerely,

    Pal Bence Bodo
    Student
    Budapest Business School
    Attached Files

  • #2
    How do I test the homogenous causality(HC) hypothesis and the heterogenous non-causality(HENC) hypothesis in Stata? Can I specify it using the xtgcause command? In other words, does the xtgcause command implement the HC and HENC developed by Dumitrescu and Hurlin(2012)? If not, is there any other command to test the HC and HENC hypothesis in Stata? I'm using panel data. Anything will help. Thanks!

    Comment

    Working...
    X