Return to Answer

added plot

Source Link

edited May 29, 2014 at 1:17

Well, in R you could try this:

library(MASS)
boxcox(y~x)
plot(1/y^2~x) # since the profile likelihood has a maximum near 2

enter image description here

But it really depends on what you mean by 'better fit to the data'

Well, in R you could try this:

library(MASS)
boxcox(y~x)
plot(1/y^2~x) # since the profile likelihood has a maximum near 2

But it really depends on what you mean by 'better fit to the data'

Well, in R you could try this:

library(MASS)
boxcox(y~x)
plot(1/y^2~x) # since the profile likelihood has a maximum near 2

enter image description here

But it really depends on what you mean by 'better fit to the data'

Source Link

created Nov 19, 2012 at 9:03

Well, in R you could try this:

library(MASS)
boxcox(y~x)
plot(1/y^2~x) # since the profile likelihood has a maximum near 2

But it really depends on what you mean by 'better fit to the data'