#172 conflictive Chi-Sq. test results for tobit residuals

closed-invalid
6
2012-09-30
2012-09-30
No

when i saved the residuals of a tobit model, and looked at the normality through "frequency distrubution" option, i've realized that the Chi-Sq, test result in "model output" window is different from which in the "frequency distrubution" window. please look at the picture enclosed...

Discussion

  • Anonymous - 2012-09-30

    conflictive results for tobit residuals

     
  • Anonymous - 2012-09-30
    • summary: conflictive Chi-Sq. results for tobit residuals --> conflictive Chi-Sq. test results for tobit residuals
     
  • Anonymous - 2012-09-30
    • priority: 5 --> 6
     
  • Allin Cottrell

    Allin Cottrell - 2012-09-30
    • assigned_to: nobody --> allin
    • labels: --> statistical output
    • status: open --> closed-invalid
     
  • Allin Cottrell

    Allin Cottrell - 2012-09-30

    The test shown in the model output window is correct: it takes into
    account the truncation inherent in the Tobit specification. Naturally,
    the standard normality test gives a different result, firmly rejecting
    normality, since that test doesn't know anything about the truncation.

     

Log in to post a comment.

Get latest updates about Open Source Projects, Conferences and News.

Sign up for the SourceForge newsletter:





No, thanks