search for: numericaly

...I know i can estimate the significance of smooth terms with chi.sq & p.value. In this example, i now that s(x1) is the most significant follow by s(x2) and s(x0). also, i know the 3 give a good contribution in my model and i have a very good model (dev expl = 97%) => But how can i estimate numericaly the contribution of each smooth against the others. In others words, is there a way to quantify this significance like a percentage of how the model is improved by each of my predictors? I hope it's ~ clear (i'm french!) thanks. [[alternative HTML version deleted]]

Hello, i got an warning message in the following code: f<-1:100 t<-1:100 b<-100 ll2 <- function(b,f,t) { t<-cumsum(t) tn<-t[length(t)] i<-seq(along=f) s1<-(tn*exp(-b*tn)*sum(f[i]))/(1-exp(-b*tn)) s2<-sum((f[i]*(t[i]*exp(-b*t[i])-t[i-1]*exp(b*t[i-1])))/(exp(-b*t[i-1])-exp(-b*t[i]))) s1-s2 } ll2(b,f,t) i think, the problem here is, that t[0] doesn't

...should only estimated signifiance of smooth terms by chi.sq?.Is there a way to link both information (p.value and chi.sq)? I have read an article where chi.sq was interpreted like residual deviance (reduction in deviance by each smooth). Can i do something like that in my case? How can i estimate numericaly the contribution of each smooth against the others. In others words, is there a way to quantify this significance like a percentage of how the model is improved by each of my smooth? Last question, using GAM with default, should i look at R-sq rather than Deviance explain, or both? I hope it'...