Announcement

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

  • Panel data analysis with period averages and dummy variables

    Dear all,
    I'm a beginner in Stata. This is my first post and I really hope to get some help from you.
    I examine the relationship between economic freedom and growth using panel data. My professor suggests that I should refer a key paper (M. Bengoa, B. Sanchez-Robles 2003) which is attached below. Here's their regression model:

    Real GDP per capitai,t = y0 interception + y1Zi,t + vi,t

    Zi,t is a set of variables that may affect economic growth. i indexes the 18 Latin American countries and t denotes the same time dimension as above (averages over 5 years from 1970 to 1999).
    I have tried to read Stata tutorials and watched some on Youtube but I still don't know how to get the result in the pic below. How to run 6 regressors of 5-year averages and get coefficient for each one. How to create a dummy variable for a specific regressor? Can I do it with one imported data set using Stata?
    Click image for larger version

Name:	table 3.png
Views:	1
Size:	127.8 KB
ID:	1452578


    I will be very grateful if you can show the Stata code in details to get a result like the above. I have attached the file of key paper and the data of my research as well in case you want to work on it. For my research, I replace Secondary and Primary enrolment with Human Development Index, and replace Public consumption with Public debt.
    Thank you in advance.
    Regards,
    Cuong Nguyen
    Attached Files

  • #2
    Cuong:
    queries like this one are often left unreplied, as they suspiciously sounds like "please, do research on my behalf studying all the stuff beforehand".
    Moreover, if what you're seeking help for is a classwork/homework, please note that this kind of queries are, at best, not encouraged on this forum (see the FAQ).

    From the table you posted, it seems that authors performed a linear panel data regression with -re- specification (as per Hausman test p-value), but reported -fe- results for sensitivity analysis purposes.
    Hence, in order to increase your chances of getting helpful replies, you should take a look at -xtreg- entry in Stata .pdf manual, try some codes yourself, come back to the list sharing what you typed and what Stata gave you back.

    Eventually, most regular contributors of this forum (me too) do not download spreadsheet from unknown sources due to the risk of active contents.
    Kind regards,
    Carlo
    (StataNow 18.5)

    Comment


    • #3
      Dear Mr. Lazzaro
      Thanks for your help. I've made much progress with my research.

      Comment


      • #4
        Cuong:
        happy with reading about your progress.
        Please call me Carlo, as many on (and many more off) the list do. Thanks.
        Kind regards,
        Carlo
        (StataNow 18.5)

        Comment

        Working...
        X