0

I have a data set where observations come from highly distinct groups. Each group may have a wildly different distribution, so I am trying to find the best distribution using fitdist from fitdistrplus, then use gamlssML from the gamlss package to find the best parameters.

My issue is with transforming the data after this step. For some of the distributions, like the Box-Cox t, I can find the equation for normalizing the data using the BCT coefficients, but for many of these distributions I cannot.

Does gamlss have a function that normalizes the data after fitting? Their documentation only provides the transformations for a small number of distributions https://www.gamlss.com/wp-content/uploads/2018/01/DistributionsForModellingLocationScaleandShape.pdf

Thanks a lot

Jage
  • 453
  • 2
  • 9

1 Answers1

1

The normalised data values (for any distribution) are exactly equal to the residuals from a gamlss fit,

m1 <- gamlss()

which can be accessed by

residuals(m1) or

m1$residuals

Robert
  • 186
  • 2
  • Thank you very much. I have been transforming my data according to page 189 of this resource https://www.gamlss.com/wp-content/uploads/2018/01/DistributionsForModellingLocationScaleandShape.pdf. But the values from this procedure are different from the residuals of gamlssML() function. Any idea why? – Jage Oct 04 '22 at 21:54
  • Are these normalized values gaussian normalized? Perhaps that is why? The transformation I have been using produces a truncated T distribution. – Jage Oct 04 '22 at 22:09
  • Indeed, the residuals are on a normal distribution. Thank you for your help. As a follow up, do you know how I can transform back to the previous scale? – Jage Oct 06 '22 at 21:11