144 lines
5.1 KiB
R
144 lines
5.1 KiB
R
# Any necessary setup
|
|
library(rpart)
|
|
options(na.action="na.omit")
|
|
options(digits=4) # to match earlier output
|
|
set.seed(1234)
|
|
|
|
mystate <- data.frame(state.x77, region=factor(state.region))
|
|
names(mystate) <- c("population","income" , "illiteracy","life" ,
|
|
"murder", "hs.grad", "frost", "area", "region")
|
|
#
|
|
# Test out the "user mode" functions, with an anova variant
|
|
#
|
|
|
|
# The 'evaluation' function. Called once per node.
|
|
# Produce a label (1 or more elements long) for labeling each node,
|
|
# and a deviance. The latter is
|
|
# - of length 1
|
|
# - equal to 0 if the node is "pure" in some sense (unsplittable)
|
|
# - does not need to be a deviance: any measure that gets larger
|
|
# as the node is less acceptable is fine.
|
|
# - the measure underlies cost-complexity pruning, however
|
|
temp1 <- function(y, wt, parms) {
|
|
wmean <- sum(y*wt)/sum(wt)
|
|
rss <- sum(wt*(y-wmean)^2)
|
|
list(label= wmean, deviance=rss)
|
|
}
|
|
|
|
# The split function, where most of the work occurs.
|
|
# Called once per split variable per node.
|
|
# If continuous=T
|
|
# The actual x variable is ordered
|
|
# y is supplied in the sort order of x, with no missings,
|
|
# return two vectors of length (n-1):
|
|
# goodness = goodness of the split, larger numbers are better.
|
|
# 0 = couldn't find any worthwhile split
|
|
# the ith value of goodness evaluates splitting obs 1:i vs (i+1):n
|
|
# direction= -1 = send "y< cutpoint" to the left side of the tree
|
|
# 1 = send "y< cutpoint" to the right
|
|
# this is not a big deal, but making larger "mean y's" move towards
|
|
# the right of the tree, as we do here, seems to make it easier to
|
|
# read
|
|
# If continuos=F, x is a set of integers defining the groups for an
|
|
# unordered predictor. In this case:
|
|
# direction = a vector of length m= "# groups". It asserts that the
|
|
# best split can be found by lining the groups up in this order
|
|
# and going from left to right, so that only m-1 splits need to
|
|
# be evaluated rather than 2^(m-1)
|
|
# goodness = m-1 values, as before.
|
|
#
|
|
# The reason for returning a vector of goodness is that the C routine
|
|
# enforces the "minbucket" constraint. It selects the best return value
|
|
# that is not too close to an edge.
|
|
temp2 <- function(y, wt, x, parms, continuous) {
|
|
# Center y
|
|
n <- length(y)
|
|
y <- y- sum(y*wt)/sum(wt)
|
|
|
|
if (continuous) {
|
|
# continuous x variable
|
|
temp <- cumsum(y*wt)[-n]
|
|
|
|
left.wt <- cumsum(wt)[-n]
|
|
right.wt <- sum(wt) - left.wt
|
|
lmean <- temp/left.wt
|
|
rmean <- -temp/right.wt
|
|
goodness <- (left.wt*lmean^2 + right.wt*rmean^2)/sum(wt*y^2)
|
|
list(goodness= goodness, direction=sign(lmean))
|
|
}
|
|
else {
|
|
# Categorical X variable
|
|
ux <- sort(unique(x))
|
|
wtsum <- tapply(wt, x, sum)
|
|
ysum <- tapply(y*wt, x, sum)
|
|
means <- ysum/wtsum
|
|
|
|
# For anova splits, we can order the categories by their means
|
|
# then use the same code as for a non-categorical
|
|
ord <- order(means)
|
|
n <- length(ord)
|
|
temp <- cumsum(ysum[ord])[-n]
|
|
left.wt <- cumsum(wtsum[ord])[-n]
|
|
right.wt <- sum(wt) - left.wt
|
|
lmean <- temp/left.wt
|
|
rmean <- -temp/right.wt
|
|
list(goodness= (left.wt*lmean^2 + right.wt*rmean^2)/sum(wt*y^2),
|
|
direction = ux[ord])
|
|
}
|
|
}
|
|
|
|
# The init function:
|
|
# fix up y to deal with offsets
|
|
# return a dummy parms list
|
|
# numresp is the number of values produced by the eval routine's "label"
|
|
# numy is the number of columns for y
|
|
# summary is a function used to print one line in summary.rpart
|
|
# In general, this function would also check for bad data, see rpart.poisson
|
|
# for instace.
|
|
temp3 <- function(y, offset, parms, wt) {
|
|
if (!is.null(offset)) y <- y-offset
|
|
list(y=y, parms=0, numresp=1, numy=1,
|
|
summary= function(yval, dev, wt, ylevel, digits ) {
|
|
paste(" mean=", format(signif(yval, digits)),
|
|
", MSE=" , format(signif(dev/wt, digits)),
|
|
sep='')
|
|
})
|
|
}
|
|
|
|
|
|
alist <- list(eval=temp1, split=temp2, init=temp3)
|
|
|
|
fit1 <- rpart(income ~population +illiteracy + murder + hs.grad + region,
|
|
mystate, control=rpart.control(minsplit=10, xval=0),
|
|
method=alist)
|
|
|
|
fit2 <- rpart(income ~population +illiteracy + murder + hs.grad + region,
|
|
mystate, control=rpart.control(minsplit=10, xval=0),
|
|
method='anova')
|
|
|
|
# Other than their call statement, and a longer "functions" component in
|
|
# fit1, fit1 and fit2 should be identical.
|
|
all.equal(fit1$frame, fit2$frame)
|
|
all.equal(fit1$splits, fit2$splits)
|
|
all.equal(fit1$csplit, fit2$csplit)
|
|
all.equal(fit1$where, fit2$where)
|
|
all.equal(fit1$cptable, fit2$cptable)
|
|
|
|
# Now try xpred on it
|
|
xvtemp <- rep(1:5, length=50)
|
|
xp1 <- xpred.rpart(fit1, xval=xvtemp)
|
|
xp2 <- xpred.rpart(fit2, xval=xvtemp)
|
|
aeq <- function(x,y) all.equal(as.vector(x), as.vector(y))
|
|
aeq(xp1, xp2)
|
|
|
|
fit3 <- rpart(income ~population +illiteracy + murder + hs.grad + region,
|
|
mystate, control=rpart.control(minsplit=10, xval=xvtemp),
|
|
method='anova')
|
|
zz <- apply((mystate$income - xp1)^2,2, sum)
|
|
aeq(zz/fit1$frame$dev[1], fit3$cptable[,4]) #reproduce xerror
|
|
|
|
zz2 <- sweep((mystate$income-xp1)^2,2, zz/nrow(xp1))
|
|
zz2 <- sqrt(apply(zz2^2, 2, sum))/ fit1$frame$dev[1]
|
|
aeq(zz2, fit3$cptable[,5]) #reproduce se(xerror)
|
|
|