The Kolmogorov-Smirnov P-values of the comparisons between the observed importance under the null hypothesis and a theoretical gaussian distributionĪ matrix with the importance scores for each permutation sample in each column The permutation test P-values ranking in decreasing order considering an approximate gaussian distribution under the null hypothesis The permutation test P-values ranking in decreasing order Testing weither the importance score is null or not. The number of permutation samples to consider for the permutation test I use the formatC function to line up the values of the i counter variable in the display. In R you cant directly compare an object to NULL using the '' operator, so you must use the is.null function. Minbucket: the minimum number of observations in any terminal node cp:Ĭomplexity parameter (Any split that does not decrease the overall lack of fit by a factor ofĬp is not attempted) maxdepth: the maximum depth of any node of the final tree, withĪ list containing the number of parallel computing arguments: The number of workers, the type of parallelization to achieve. Counter variable i is explicitly coerced to an integer using the as.integer function because the default numeric type in R is type double. The improper survival tree parameters: a list of options that control details of the rpart algorithm. The number of Bagging samples to consider The splitting method: either "R2" for the proposed pseudo-R2 criterion or "LR" for the adjusted Logrank criterion The importance score to consider: either IIS, DIIS or DEPTH The names of independant variables acting on the survival of the susceptible population The names of independant variables acting on the non-susceptible population (the plateau) Variable selection using the permutation test on several scores of importance: IIS, DIIS and DEPTH.Ī vector of the names of the two variables of interest (the time-to-event is follow by the event indicator)
0 kommentar(er)
