Announcement

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

  • STATA exercisea

    Hi, everybody!
    could anyone help with this two tasks?
    Please find the data for stata here https://drive.google.com/open?id=1l1...nQ7IM6UPvjwnNf
    The pictufe of table 1-2 at the end of doc,

    1 -
    STATA exercise:
    In this Stata problem, you will use the data set “PanelB.dta” on Courseworks to replicate the
    results from the policy paper “Does Abolishing User Fees Lead to Improved Health Status? Evidence from
    Post-Apartheid South Africa” used and discussed in class. We discussed in class the results in Panel A of Table 2.
    We examined a difference in differences regression comparing the WAZ-score of children younger than 4 in 1998
    to that of children younger than 4 in 1993. We then compared the results of our regression to the results reported
    in the paper.
    In this problem you will have to replicate the results in Panel B of Table 2. The panel reports the results of a
    difference in differences regression comparing the WAZ-score of children who were younger than 4 in 1993, and
    between ages of 5 and 8 in 1998. The data contains only the observations you would need for Panel B. You will
    still have to exclude the WAZ-score smaller than -6 and larger than 5, as the author does.
    (a)
    (4 points)
    Run a regression for a difference-in-differences estimation, excluding the observations with
    WAZ-score smaller than -6 and larger than 5. Show the STATA command and its output.
    (b)
    (4 points)
    Write down a 3x3 table like Table 2 in the paper, with column headers labeled “High”, “Low”,
    and “Difference”, and row headers labeled “Post-reform”, “Pre-reform”, and “Difference”. Place your
    regression coefficients, with their standard errors below them in parentheses, in each corresponding cell.
    (Note that you will only fill out 4 cells out of 9 cells in the table.) Your standard errors will be different, but
    the estimates should be identical to those in the paper.
    (c)
    (10 points)
    Using the appropriate lincom
    commands, fill out the remaining cells in Panel B with estimates
    and their standard errors in parentheses below them. Again, your standard errors will be different, but the
    estimates should be identical to those in the paper. Show your STATA commands and your output. Shade or
    otherwise identify which cells in the table you are filling out in this part.
    ***
    2
    STATA exercise:
    Download the STATA dataset “ECLS
    Spring2018.dta” from Coursework and open it in
    Stata. The data come from Early Childhood Longitudinal Study, a nationally representative sample that collects
    information from children, their families, their teachers, and their schools.
    (a)
    (4 points)
    How does a child’s household income, whether the family is below the poverty line, and a child’s
    mother’s years of education affect the child’s performance on a math score? Show the STATA command and
    its output.
    (b)
    (6 points) Interpret each of the coefficients of the regression in the previous part.
    (c)
    (2 points)
    Using the regression in part a), what is the effect of income on predicted math scores for a child
    whose mother graduated high school (that is, who has 12 years of schooling
    )?
    (d)
    (2 points)
    Using the regression in part a), what is the effect of income on predicted math scores for a child
    whose mother has a 2-year master’s degree (that is, who has 18 years of schooling
    )?
    (e)
    (2 points)
    Compare the answers to the two questions above. Does your comparison surprise you? Why or
    why not?
    (f)
    (2 points)
    Do you think the comparison in the previous part reflects reality? If not, how would you modify
    the regression to account for this discrepancy?
    (g)
    (6 points)
    Run a regression which allows you to give different answers to the questions c) and d). Include the
    variable
    poverty
    . Show the STATA command and its output. Interpret the coefficients of this regression.
    (h)
    (6 points)
    In the regression in part g), what is the effect of income on predicted math scores for a child
    whose mother graduated high school? Is this effect statistically significant at a 1% significance level? Show
    the STATA command and its output.
    (i)
    (6 points)
    In the regression in part g), what is the effect of income on predicted math scores for a child
    whose mother has a 2-year master’s degree? Is this effect statistically significant at a 1% significance level?
    Show the STATA command and its output.


    Click image for larger version

Name:	2018-06-23 04.18.35.png
Views:	1
Size:	77.7 KB
ID:	1450259
    To use Google Drive, you need to use a supported web browser. Learn what a browser is, which browser you use, and how to update to the latest version. Option 1: Find the update for your browser To

  • #2
    Guest,

    Per the FAQ, we aren't encouraged to help with homework questions, because

    However, coursework that will be graded raises some very different issues. To start at the sharp end, many institutions have strict rules about plagiarism and collusion. So, consulting others and receiving help from others may place students in difficult situations, with very possible serious risk to their grades and their careers. It is thus not at all in students' best interests.

    Most of all, homework is in essence set as fair challenges to students with the expectation that they can work their own way towards answers and will benefit from doing so. All of us have been students at some point, and many of us are academics with teaching roles. Long-term members here tend to agree strongly that the ethos of being helpful stops short of providing help on whatever is obviously homework. Institutions will have their own support rules and set-ups (faculty office hours, teaching assistants, whatever), but out-sourcing homework support to internet forums is not part of any policy we wish to support.
    Short answers is that you have to learn to do it yourself - but your teaching assistants are paid to help you understand how to do it, and you should hit them and/or the professor up.

    I will add that with regards to 1B, you can probably get by with just copy-pasting results from the Stata window to Excel. The format is preserved well enough. However, in the long term, you want to learn about user-written commands like -estout-, that you have to install manually (kind of like packages in R, except the functionality of base Stata is much greater than base R) but that are vetted by the Stata community. You'd type

    ssc install estout
    Last edited by sladmin; 25 Jun 2018, 08:15. Reason: Anonymize original poster
    Be aware that it can be very hard to answer a question without sample data. You can use the dataex command for this. Type help dataex at the command line.

    When presenting code or results, please use the code delimiters format them. Use the # button on the formatting toolbar, between the " (double quote) and <> buttons.

    Comment


    • #3
      I tend to be supportive when someone, after having tried hard to solve a tricky problem, even in a homework, demands specific help in order to overcome a given obstacle.

      But it doesn’t seem to be the case here.

      It is absolutely out of the purpose of this Forum - as well as whatsoever place, I daresay - to provide solutions for a full homework. After all, I can hardly envisage other aim of attending a course if not learning, and this shall be demonstrated by dutifully fulfilling the tasks.
      Best regards,

      Marcos

      Comment


      • #4
        Guest:
        it is hard to be symphatetic with someone who asks to download the data, perform her/his homework and then report back to her/him with the results.
        Last edited by sladmin; 25 Jun 2018, 08:16. Reason: Anonymize original poster
        Kind regards,
        Carlo
        (Stata 18.0 SE)

        Comment


        • #5
          I guess that I don't need to endorse people in alerting Guest to the FAQ.

          There may be extra credit for explaining to the teacher that the spelling STATA has been non-standard for more than 30 years.
          Last edited by sladmin; 25 Jun 2018, 08:16. Reason: Anonymize original poster

          Comment

          Working...
          X