# Transformations: Function for Data Transformations In party: A Laboratory for Recursive Partytioning

## Description

Transformations of Response or Input Variables

## Usage

 ```1 2 3 4``` ```ptrafo(data, numeric_trafo = id_trafo, factor_trafo = ff_trafo, ordered_trafo = of_trafo, surv_trafo = logrank_trafo, var_trafo = NULL) ff_trafo(x) ```

## Arguments

 `data` an object of class `data.frame`. `numeric_trafo` a function to by applied to `numeric` elements of `data` returning a matrix with `nrow(data)` rows and an arbitrary number of columns. `ordered_trafo` a function to by applied to `ordered` elements of `data` returning a matrix with `nrow(data)` rows and an arbitrary number of columns (usually some scores). `factor_trafo` a function to by applied to `factor` elements of `data` returning a matrix with `nrow(data)` rows and an arbitrary number of columns (usually a dummy or contrast matrix). `surv_trafo` a function to by applied to elements of class `Surv` of `data` returning a matrix with `nrow(data)` rows and an arbitrary number of columns. `var_trafo` an optional named list of functions to be applied to the corresponding variables in `data`. `x` a factor

## Details

`trafo` applies its arguments to the elements of `data` according to the classes of the elements. See `Transformations` for more documentation and examples.

In the presence of missing values, one needs to make sure that all user-supplied functions deal with that.

## Value

A named matrix with `nrow(data)` rows and arbitrary number of columns.

## Examples

 ```1 2 3 4 5 6``` ``` ### rank a variable ptrafo(data.frame(y = 1:20), numeric_trafo = function(x) rank(x, na.last = "keep")) ### dummy coding of a factor ptrafo(data.frame(y = gl(3, 9))) ```

party documentation built on March 4, 2021, 1:06 a.m.