Description Usage Arguments Value Author(s) References See Also Examples
Evaluates the cross-validation criterion of Hazelton and Turlach (2009) for the bivariate case.
1 | cv.score.mv(y, Sigma, H, gamma, K=5, verb=FALSE)
|
y |
matrix with two columns containing the observed values. |
Sigma |
the variance-covariance matrix of the contaminating (normal) distribution. |
H |
the matrix of smoothing parameter to be used. |
gamma |
vector of values from which a suitable value is to be selected |
K |
number of folds to be used if |
verb |
logical; if |
A vector containing the cross-validation criterion evaluated at the
values given in gamma
.
Martin L Hazelton m.hazelton@massey.ac.nz
Hazelton, M.L. and Turlach, B.A. (2009). Nonparametric density deconvolution by weighted kernel estimators, Statistics and Computing 19(3): 217–228. http://dx.doi.org/10.1007/s11222-008-9086-7.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 | library(ks)
Age <- framingham[,2]
Age.lim.2 <- 56
SBP1.A <- framingham[Age>=Age.lim.2,3] # SBP, measure 1, Exam 2
SBP2.A <- framingham[Age>=Age.lim.2,4] # SBP, measure 2, Exam 2
SBP1.B <- framingham[Age>=Age.lim.2,5] # SBP, measure 1, Exam 3
SBP2.B <- framingham[Age>=Age.lim.2,6] # SBP, measure 2, Exam 3
sigma.fram.A <- sd(SBP1.A-SBP2.A)
sigma.fram.B <- sd(SBP1.B-SBP2.B)
Sigma.fram <- diag(c(sigma.fram.A,sigma.fram.B))^2
SBP.A <- SBP1.A
SBP.B <- SBP1.B
SBP.bi <- cbind(SBP.A,SBP.B)
H.fram <- Hpi(SBP.bi)
gamma.ridge <- exp(seq(from=-2, to=3, length=17))
set.seed(100813)
cv <- cv.score.mv(SBP.bi, Sigma=Sigma.fram, H=H.fram,
gamma=gamma.ridge, verb=TRUE)
plot(log(gamma.ridge), cv, type="b")
tmp <- getmin(log(gamma.ridge), cv, which="r")
abline(v=tmp$xmin)
set.seed(100813)
cv <- cv.score.mv(SBP.bi, Sigma=Sigma.fram, H=H.fram,
gamma=gamma.ridge, verb=TRUE, K=10)
plot(log(gamma.ridge), cv, type="b")
tmp <- getmin(log(gamma.ridge), cv, which="r")
abline(v=tmp$xmin)
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.