Regression for Non-normal data

Six Sigma – iSixSigma Forums Old Forums General Regression for Non-normal data

Viewing 3 posts - 1 through 3 (of 3 total)
  • Author
  • #32251


    I am looking at a non-normal Cycle time data (KPOV) which is skewed left. One of the KPIVs show strong correlation with the KPOV. 
    Linear regression uses the underline assumption that the data is normal. Is there a way I can map the effect of the KPIV to KPOV in this case so I can predict if reducing this KPIV significantly helps reduce the KPOV ? Can I do regression if data is not normal ?



    Why do you believe the data needs to be normal? I thinik this least squares linear fit is independent of the data form.
    There is a requirement for normality of the residuals in order to assess the model fit.
    However, I know of no reference that says the data has  any distributional requirement. Do you?
    I have searched Draper and Smith “Applied Regression Analysis” and can not find such a requirement.


    Robert Butler

    To Dave’s point-pp.22 of Draper and Smith Applied Regression Analysis, Second Edition states with respect to regression analysis “Up to this point we have made no assumptions at all that involve probability distributions. A number of specified algebraic calculations have been made and that is all.”
    On that same page and the one following they are very clear about the fact that assumptions about normality apply to the residuals only.

Viewing 3 posts - 1 through 3 (of 3 total)

The forum ‘General’ is closed to new topics and replies.