lda {Kira}

Linear discriminant analysis (LDA).

Description

Perform linear discriminant analysis.

Usage

lda(data, test = NA, class = NA, type = "train", 
   method = "moment", prior = NA)

Arguments

data

Data to be classified.

test

Vector with indices that will be used in 'data' as test. For type = "train", one has test = NA.

class

Vector with data classes names.

type

Type of type:
"train" - data training (default), or
"test" - classifies the data of the vector "test".

method

Classification method:
"mle" to MLEs,
"mve" to use cov.mv,
"moment" (default) for standard mean and variance estimators, or
"t" for robust estimates based on the t distribution.

prior

Probabilities of occurrence of classes. If not specified, it will take the proportions of the classes. If specified, probabilities must follow the order of factor levels.

Value

predict

The classified factors of the set.

Author(s)

Paulo Cesar Ossani

References

Rencher, A. C. Methods of multivariate analysis. 2th. ed. New York: J.Wiley, 2002. 708 p.

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

Mingoti, S. A. Analise de dados atraves de metodos de estatistica multivariada: uma abordagem aplicada. Belo Horizonte: UFMG, 2005. 297 p.

Ferreira, D. F. Estatistica Multivariada. 2a ed. revisada e ampliada. Lavras: Editora UFLA, 2011. 676 p.

See Also

plot_curve and results

Examples

data(iris) # data set

data  <- iris
names <- colnames(data)
colnames(data) <- c(names[1:4],"class")

#### Start - hold out validation method ####
dat.sample = sample(2, nrow(data), replace = TRUE, prob = c(0.7,0.3))
data.train = data[dat.sample == 1,] # training data set
data.test  = data[dat.sample == 2,] # test data set
class.train = as.factor(data.train$class) # class names of the training data set
class.test  = as.factor(data.test$class)  # class names of the test data set
#### End - hold out validation method ####

r <- (ncol(data) - 1)
class <- data[,c(r+1)] # classes names

## Data training example
res <- lda(data = data[,1:r], test = NA, class = class, 
           type = "train", method = "moment", prior = NA)

resp <- results(orig.class = class, predict = res$predict)

message("Mean squared error:"); resp$mse
message("Mean absolute error:"); resp$mae
message("Relative absolute error:"); resp$rae
message("Confusion matrix:"); resp$conf.mtx  
message("Hit rate: ", resp$rate.hits)
message("Error rate: ", resp$rate.error)
message("Number of correct instances: ", resp$num.hits)
message("Number of wrong instances: ", resp$num.error)
message("Kappa coefficient: ", resp$kappa)
message("General results of the classes:"); resp$res.class 


## Data test example
class.table <- table(class) # table with the number of elements per class
prior <- as.double(class.table/sum(class.table))
test = as.integer(rownames(data.test)) # test data index

res <- lda(data = data[,1:r], test = test, class = class, 
           type = "test", method = "mle", prior = prior)

resp <- results(orig.class = class.test, predict = res$predict)

message("Mean squared error:"); resp$mse
message("Mean absolute error:"); resp$mae
message("Relative absolute error:"); resp$rae
message("Confusion matrix: "); resp$conf.mtx  
message("Hit rate: ", resp$rate.hits)
message("Error rate: ", resp$rate.error)
message("Number of correct instances: ", resp$num.hits)
message("Number of wrong instances: ", resp$num.error)
message("Kappa coefficient: ", resp$kappa)
message("General results of the classes:"); resp$res.class