Announcement

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

  • Residuals in rolling regression

    Hello Dear Statalist,
    I have an unbalanced panel data, that contains the mutual funds, their returns and the name of the managers/team that manage the funds.I have monthly data for 20 years for more than 3500 funds I have to investigate the riskiness and performance of single vs team managed funds.
    I have to run rolling regression over the past 12 consecutive month in order to estimate the fund betas and alphas. So I will obtain the systematic risk of funds. I am using the following code:

    rolling, window(12) reject(e(N)<12) saving(stats, replace): reg mretrf mktrf, vce(robust)

    My problem is that I don't know which command to use to obtain the standard deviation of fund i's residual fund return from this model. Could someone provide me the needed code? Thanks in advance!


  • #2
    Anybody?

    Comment


    • #3
      IS my question not clear or the explanation is not sufficient?

      Comment


      • #4
        Dilyana: The way Statalist works is hard to predict, even when you've been a member for a long while.

        If a question is a little bit unclear, people tend to say that, and ask you to fill in a gap.

        If no-one wants to answer, then usually no-one will answer.

        Either way, bumping a thread, which you do often, is usually a bad idea. It won't make a question clearer. No one will answer more quickly because you ask repeatedly. At most, wait one day before reminding people once of your question.

        The good news about Statalist is that you can get good answers quickly for free when someone competent likes your question and can answer it. The bad news is that no one is obliged to answer anything and can choose any reason they like for not answering, from being busy onwards.

        If that doesn't work for you, you need to read the documentation and/or take a course in Stata and learn more for yourself, or hire an assistant.
        Last edited by Nick Cox; 23 Jun 2015, 09:30.

        Comment


        • #5
          Ok Nick I am sorry about that.I have read a lot of posts but I didn't find the solution of my problem. The only post that is similar to my problem is http://www.statalist.org/forums/foru...ing-regression but I can not understand the code. Is it actually possible to obtain residuals from rolling regression

          Comment


          • #6
            Every regression generates as many residuals as there are data points. For points 1 to 12 there are 12 residuals. For points 2 to 13, 12 more, and so forth.

            Comment

            Working...
            X