+ Reply to Thread
Results 1 to 1 of 1

Thread: When should I report heteroskedasticity robust standard errors?

  1. #1
    Points: 263, Level: 5
    Level completed: 26%, Points required for next Level: 37

    Posts
    2
    Thanks
    0
    Thanked 0 Times in 0 Posts

    When should I report heteroskedasticity robust standard errors?




    With regard to a master's thesis in economics I am working on I am running six regressions (see attachment).
    I find evidence for heteroskedasticity (by White's test) in one of the six regression outputs (regression 2 in attached file).

    The question is: Should I report robust standard errors for that specific output?
    Or should I report robust standard errors for all six regression (allthough there is no evidence for heteroskedasticity in five of them).

    At the moment I have for the sake of consistency reported non-robust standard errors in all six regression outputs.


    With robust errors, this is the result of the specific regression:

    Dependent variable: Linux
    Heteroskedasticity-robust standard errors, variant HC1

    coefficient std. error t-ratio p-value
    -------------------------------------------------------------
    const 0.0547160 0.0168694 3.244 0.0016 ***
    Piracyrate2010 -0.0216238 0.00991563 -2.181 0.0315 **
    l_GDP_capita -0.00311412 0.00125895 -2.474 0.0150 **
    AP -0.00626468 0.00175146 -3.577 0.0005 ***
    CEE 0.00548235 0.00494118 1.110 0.2698
    MEA -0.00653328 0.00163738 -3.990 0.0001 ***
    Attached Images

+ Reply to Thread

           




Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts






Advertise on Talk Stats