Classify Multivariate Observations by Linear Discrimination

Classify multivariate observations in conjunction with lda, and also project data onto the linear discriminants.

# S3 method for lda
predict(object, newdata, prior = object$prior, dimen,
        method = c("plug-in", "predictive", "debiased"), ...)

Arguments

object	object of class "lda"
newdata	data frame of cases to be classified or, if object has a formula, a data frame with columns of the same names as the variables used. A vector will be interpreted as a row vector. If newdata is missing, an attempt will be made to retrieve the data used to fit the lda object.
prior	The prior probabilities of the classes, by default the proportions in the training set or what was set in the call to lda.
dimen	the dimension of the space to be used. If this is less than min(p, ng-1), only the first dimen discriminant components are used (except for method="predictive"), and only those dimensions are returned in x.
method	This determines how the parameter estimation is handled. With "plug-in" (the default) the usual unbiased parameter estimates are used and assumed to be correct. With "debiased" an unbiased estimator of the log posterior probabilities is used, and with "predictive" the parameter estimates are integrated out using a vague prior.
...	arguments based from or to other methods

Value

a list with components

class

The MAP classification (a factor)

posterior

posterior probabilities for the classes

x

the scores of test cases on up to dimen discriminant variables

Details

This function is a method for the generic function predict() for class "lda". It can be invoked by calling predict(x) for an object x of the appropriate class, or directly by calling predict.lda(x) regardless of the class of the object.

Missing values in newdata are handled by returning NA if the linear discriminants cannot be evaluated. If newdata is omitted and the na.action of the fit omitted cases, these will be omitted on the prediction.

This version centres the linear discriminants so that the weighted mean (weighted by prior) of the group centroids is at the origin.

References

Venables, W. N. and Ripley, B. D. (2002) Modern Applied Statistics with S. Fourth edition. Springer.

Ripley, B. D. (1996) Pattern Recognition and Neural Networks. Cambridge University Press.

See also

lda, qda, predict.qda

Examples

tr <- sample(1:50, 25)
train <- rbind(iris3[tr,,1], iris3[tr,,2], iris3[tr,,3])
test <- rbind(iris3[-tr,,1], iris3[-tr,,2], iris3[-tr,,3])
cl <- factor(c(rep("s",25), rep("c",25), rep("v",25)))
z <- lda(train, cl)
predict(z, test)$class
#>  [1] s s s s s s s s s s s s s s s s s s s s s s s s s c c c c c c c c c c v c c
#> [39] c c c c c c c c c c c c v v v v v v v v v c v v v v v v v v v v v v v v v
#> Levels: c s v