vtreat (version 1.0.1)

mkCrossFrameCExperiment: Run categorical cross-frame experiment.

Description

Builds a designTreatmentsC treatment plan and a data frame prepared from dframe that is "cross" in the sense each row is treated using a treatment plan built from a subset of dframe disjoint from the given row. The goal is to try to and supply a method of breaking nested model bias other than splitting into calibration, training, test sets.

Usage

mkCrossFrameCExperiment(dframe, varlist, outcomename, outcometarget, ...,
  weights = c(), minFraction = 0.02, smFactor = 0, rareCount = 0,
  rareSig = 1, collarProb = 0, codeRestriction = NULL,
  customCoders = NULL, scale = FALSE, doCollar = FALSE,
  splitFunction = NULL, ncross = 3, forceSplit = FALSE,
  catScaling = FALSE, parallelCluster = NULL)

Arguments

dframe

Data frame to learn treatments from (training data), must have at least 1 row.

varlist

Names of columns to treat (effective variables).

outcomename

Name of column holding outcome variable. dframe[[outcomename]] must be only finite non-missing values.

outcometarget

Value/level of outcome to be considered "success", and there must be a cut such that dframe[[outcomename]]==outcometarget at least twice and dframe[[outcomename]]!=outcometarget at least twice.

...

no additional arguments, declared to forced named binding of later arguments

weights

optional training weights for each row

minFraction

optional minimum frequency a categorical level must have to be converted to an indicator column.

smFactor

optional smoothing factor for impact coding models.

rareCount

optional integer, allow levels with this count or below to be pooled into a shared rare-level. Defaults to 0 or off.

rareSig

optional numeric, suppress levels from pooling at this significance value greater. Defaults to NULL or off.

collarProb

what fraction of the data (pseudo-probability) to collar data at if doCollar is set during prepare.

codeRestriction

what types of variables to produce (character array of level codes, NULL means no restiction).

customCoders

map from code names to custom categorical variable encoding functions (please see https://github.com/WinVector/vtreat/blob/master/extras/CustomLevelCoders.md).

scale

optional if TRUE replace numeric variables with regression ("move to outcome-scale").

doCollar

optional if TRUE collar numeric variables by cutting off after a tail-probability specified by collarProb during treatment design.

splitFunction

(optional) see vtreat::buildEvalSets .

ncross

optional scalar>=2 number of cross-validation rounds to design.

forceSplit

logical, if TRUE force cross-validated significance calculatons on all variables.

catScaling

optional, if TRUE use glm() linkspace, if FALSE use lm() for scaling.

parallelCluster

(optional) a cluster object created by package parallel or package snow

Value

list with treatments and crossFrame

See Also

designTreatmentsC designTreatmentsN prepare

Examples

# NOT RUN {
set.seed(23525)
zip <- paste('z',1:100)
N = 200
d <- data.frame(zip=sample(zip,N,replace=TRUE),
                zip2=sample(zip,20,replace=TRUE),
                y=runif(N))
del <- runif(length(zip))
names(del) <- zip
d$y <- d$y + del[d$zip2]
d$yc <- d$y>=mean(d$y)
cC <- mkCrossFrameCExperiment(d,c('zip','zip2'),'yc',TRUE,
  rareCount=2,rareSig=0.9)
cor(as.numeric(cC$crossFrame$yc),cC$crossFrame$zip_catB)  # poor
cor(as.numeric(cC$crossFrame$yc),cC$crossFrame$zip2_catB) # better
treatments <- cC$treatments
dTrainV <- cC$crossFrame

# }