Hi I'm following from p. dickman's website
Just wondering why does he recommend using a spline variable for the year of diagnosis?
I know splines are used for non-linear trends, but I don't understand why he recommends splitting it into a spline
My aim is to plot a graph to model hazard comparing males: females for a treatment = 1 over time
The problem isn't coding but the logic of why a spline was introduced
Just wondering why does he recommend using a spline variable for the year of diagnosis?
I know splines are used for non-linear trends, but I don't understand why he recommends splitting it into a spline
My aim is to plot a graph to model hazard comparing males: females for a treatment = 1 over time
The problem isn't coding but the logic of why a spline was introduced
Code:
clear all use https://pauldickman.com/data/melanoma.dta rename surv_mm survivalt stset survivalt, failure(status==1) exit(time 60.5) scale(12) //create dummy variable - treament gen treatment = 0 replace treatment = 1 if stage == 1 stpm2 treatment age stage sex, scale(hazard) df(4) eform nolog tvc(treatment sex) dftvc(3) //Graph 3: model hazard in terms of male : Female for treatment ==1 over time //Creating dummy variables for modelling gen female = sex==2 gen male= sex==1 //spline variables for year of diagnosis rename yydx opdate rcsgen opdate, df(3) gen(yearspl) orthog //look up df , orthog ****pending range temptime 0 10 51 //refit model with spline in year - WHY IS THIS DONE? stpm2 treatment age yearspl* stage sex, scale(hazard) df(5) tvc(treatment sex) dftvc(3) //Time varying effect of gender and specific for treatment = 1 //Males treatment = 1 predict hr6, hrnumerator (sex 1 treatment 1) hrdenominator (sex 2 treatment 1) ci timevar(temptime)
