Announcement

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

  • Stata regression command help

    Hi everyone!

    I am working on a project where I will be studying how subcontracting affects the ability of the US economy to recover from economic recessions. I will be using payroll by broad US industry class as my dependent variable to measure economic recovery. My control variables will be unemployment rate by US geographic region, real GDP by region, and real personal income by region. I will have IVs about levels of subcontracting by industry and by region.

    Does anyone have any thoughts on what regression command I might use in Stata to get at what I am trying to study?

    Thank you!
    Jacob

  • #2
    Jacob, this question repeats one you have posed repeatedly on Statalist over the past few weeks, none of which have resulted in advice for you.

    Your question is not about Stata per se, but rather about econometric methodology, which is what drives the choice of regression command. A question about Stata would at least start with a description of the methodology - "I want to use two stage least squares to fit a model of ..." - which suggests appropriate commands to a knowledgeable econometrician. But if you know the methodology, the Stata documentation and help command make is reasonably straightforward to find suitable commands.

    So the pool of members who might respond to you is narrow, and those who might be familiar with the sort of modeling you hope to accomplish may find your question about the appropriate command to be premature given the vague description of the model to be built.

    Realistically, Statalist is best suited for discussing questions around using Stata commands for a well-defined purpose. I think your question would be better suited to a discussion on a forum focusing on econometrics.

    Comment


    • #3
      William Lisowski is bang on here, as usual.

      Jacob: You should perhaps study some of the reasons why questions don't get answered here. https://www.statalist.org/forums/help#noanswer

      No doubt it's threads like these that don't (seem likely to) get very far that contribute to assertions like that in https://twitter.com/inandoutofcars/s...68713372262400 However, Statalist is a forum, not a helpline. StataCorp employees aside, everyone answering here makes their own judgements about what can and should be answered.

      Comment


      • #4
        My apologies, I honestly thought that this was a slight variation on the other questions which I had been asking. I was attempting to more clearly define my model and then ask if anyone had any advice on a regression command which I might be able to use. I thought this was different.

        Mr. Cox: As you can clearly see, I am new to statistical modeling and am trying my very best. I enjoy using Stata and enjoy the work that I do. If I have offended you by post multiple times, thinking that I was improving my question each time, I am very sorry. However, that being said, I really do not appreciate you comparing me to comments such as that on the Twitter thread. That was unnecessary and in my opinion quite unkind. I am a student, thing my best to understand the work that I am doing. We are all human- there is no need to attack students who are just trying their best. I wish you only the best.

        Comment


        • #5
          Furthermore, I do not believe that Stata is a pain or that people here are unhelpful. I appreciate the fact that you are here to help people.

          Comment


          • #6
            I am not offended by your posts and am not attacking you.

            I was just commenting with a quotation underlining that clearly some people don't like Statalist; in the absence of evidence I can only guess that is because in some cases we don't answer their questions, or did give them an answer they didn't like. We are not answering your questions, at least not in the way that you want; that's the only link. Let me stress that I didn't think it was you that made that Twitter post, and as it wasn't you then naturally you aren't responsible for it in any way.

            My concern is first and foremost that questions like these are not helping you -- and so the implication is that you need to look elsewhere for help. I don't know the nature of your project, but. if it's a coursework there should be either (a) support mechanisms at your institution -- faculty hours, TAs, whatever -- or (b) an expectation that you work at this yourself.

            One has only to look at your previous thread to see there https://www.statalist.org/forums/for...o-use-in-stata advice that posting the same question repeatedly is not a good idea. You describe the first post here #1 as a "slight variation" on previous posts, but I don't see it as much different.

            William Lisowski and I are just people who post here more than most. Nothing stops anyone else answering your questions if so minded, but the advice from him and me is consistent.

            Comment

            Working...
            X