compare, compare.kda.diag.cv, compare.kda.cv, compare.pda.cv {ks} | R Documentation |
Comparisons for kernel and parametric discriminant analysis.
compare(x.group, est.group) compare.kda.cv(x, x.group, bw="plugin", prior.prob=NULL, Hstart, ...) compare.kda.diag.cv(x, x.group, bw="plugin", prior.prob=NULL, ...) compare.pda.cv(x, x.group, type="quad", prior.prob=NULL)
x |
matrix of training data values |
x.group |
vector of group labels for training data |
est.group |
vector of estimated group labels |
bw |
"plugin" = plug-in, "lscv" = LSCV, "scv" = SCV |
Hstart |
(stacked) matrix of initial bandwidth matrices |
prior.prob |
vector of prior probabilities |
type |
"line" = linear discriminant, "quad" =
quadratic discriminant |
... |
other optional parameters for bandwidth selection, see
Hpi , Hlscv , Hscv |
If you have prior probabilities then set prior.prob
to these.
Otherwise the default is to use the sample proportions as
estimates of the prior probabilities.
The parametric discriminant analysers use the code from the
MASS
library namely lda
and qda
for linear and
quadratic discriminants.
The function compare
creates a comparison between the true
group labels x.group
and the estimated ones est.group
.
It returns a list with fields
cross |
cross-classification table with the rows indicating the true group and the columns the estimated group |
error |
misclassification rate (MR) |
In the case where we have test data that is independent of the
training data, then
MR = (number of points wrongly classified) / (total number of points)
In the case where we don't have independent test data e.g.
we are classifying the
training data set, then the cross validated estimate is more
appropriate. See Silverman (1986). These are implemented as for
kernel discriminant analysis as compare.kda.cv
(full bandwidth
selectors) and compare.kda.diag.cv
(for diagonal bandwidth
selectors), and compare.pda.cv
for parametric discriminant analysis.
Silverman, B. W. (1986) Data Analysis for Statistics and Data Analysis. Chapman & Hall. London.
Simonoff, J. S. (1996) Smoothing Methods in Statistics. Springer-Verlag. New York
Venables, W.N. & Ripley, B.D. (1997) Modern Applied Statistics with S-PLUS. Springer-Verlag. New York.
### bivariate example - restricted iris dataset library(MASS) data(iris) ir <- iris[,c(1,2)] ir.gr <- iris[,5] compare.kda.cv(ir, ir.gr, bw="plug-in", pilot="samse") compare.pda.cv(ir, ir.gr, type="line") compare.pda.cv(ir, ir.gr, type="quad")