3 replies to this topic

[P-A Leboeuf]

Hi,
I a new user of Phoenix winNonlin and :

 

I want to model strange and poor data. Someone proposed me to try a hockey stick model.

It just the combination of two linear model, but I want to estimate the threshold. I don't want to set by myself the threshold. I am suspecting a "grammarian" error in my text editor. Unfortunately, it is not estimated in the "theta" table and it is like the initial estimate without variation just like we freeze the parameters

 

 

test(){

covariate©

E = ((C < threshold)? (Alpha * (1 + Beta*C)) : (Alpha * (1 + Beta2*C)) )

error(EEps = 1)

observe(EObs© = E + EEps)

stparm(Alpha = tvAlpha)

stparm(Beta = tvBeta)

stparm(Beta2 = tvBeta2)

stparm(threshold= tvlthreshold)

fixef(tvAlpha = c(, 80, ))

fixef(tvBeta = c(, -1, ))

fixef(tvBeta2 = c(, -2, ))

fixef(tvlthreshold = c(,20,))

}

 

 

thanks, 

[smouksassi1]

Hi Paul,

you need to be careful on how to define the joining point between the two parts of the curve:

 

 

test(){

covariate(x)

E =  x<= threshold ? intercept :   (threshold*slope)- x* slope

error(EEps = 1)

observe(y(x) = E + EEps)

stparm(intercept = tvintercept)

stparm(slope = tvslope)

fixef(threshold = c(, 40, ))

fixef(tvintercept = c(, 1, ))

fixef(tvslope = c(, 4.5, ))

}

 

here I show how to fit a curve that start flat then it starts to go down with x see the attached phoenix project.

Attached Files

•  Paul.phxproj   349.12KB   325 downloads

[Simon Davis]

If you're not sure about the inflection/threshold point, here is a simpler log-linear model derived from PD30 in Gabrielsson and Weiner if that helps you any?

 

test(){
    covariate©
    E = m * ln(C0+ C)
    error(EEps = 1)
    observe(EObs© = E + EEps)
    fixef(m = c(, 40, ), C0 = c(, 1.6, ))
}

 

 Simon.

[P-A Leboeuf]

Thanks to both of you!

 

I found my solution in the project of M.Mouksassi

 

test(){

covariate©

E =  C<= threshold ? intercept :  (intercept+threshold*slope )  - C*slope

error(EEps = 1)

observe(y© = E + EEps)

stparm(intercept = tvintercept)

stparm(slope = tvslope)

fixef(threshold = c(, 20, ))

fixef(tvintercept = c(, 70, ))

fixef(tvslope = c(, 5, ))

}