Directional (version 4.0)

k-NN regression: k-NN regression with Euclidean or (hyper-)spherical response and or predictor variables

Description

k-NN regression with Euclidean or (hyper-)spherical response and or predictor variables.

Usage

knn.reg(xnew, y, x, k = 5, res = "eucl", type = "euclidean", estim = "arithmetic")

Arguments

xnew

The new data, new predictor variables values. A matrix with either euclidean (univariate or multivariate) or (hyper-)spherical data. If you have a circular response, say u, transform it to a unit vector via (cos(u), sin(u)). If xnew = x, you will get the fitted values.

y

The currently available data, the response variables values. A matrix with either euclidean (univariate or multivariate) or (hyper-)spherical data. If you have a circular response, say u, transform it to a unit vector via (cos(u), sin(u)).

x

The currently available data, the predictor variables values. A matrix with either euclidean (univariate or multivariate) or (hyper-)spherical data. If you have a circular response, say u, transform it to a unit vector via (cos(u), sin(u)).

k

The number of nearest neighbours, set to 5 by default. This can also be a vector with many values.

res

The type of the response variable. If it is Euclidean, set this argument equal to "res". If it is a unit vector set it to res="spher".

type

The type of distance to be used. This determines the nature of the predictor variables. This is actually the argument "method" of the distance function in R. The default value is "euclidean". If you use the Euclidean distance, the package "Rfast" is used. The "dista" function of that package is about 3 times faster than the standard built-in "dist". R has several options the type of the distance. Just type ?Rfast::Dist in R and see the methods. Any method can be given here. If you have unit vectors in general, you should put type="spher", so that the cosinus distance is calculated.

estim

Once the k observations whith the smallest distance are discovered, what should the prediction be? The arithmetic average of the corresponding y values be used estim="arithmetic" or their harmonic average estim="harmonic".

Value

A list with as many elements as the number of values of k. Each element in the list contains a matrix (or a vector in the case of Euclidean data) with the predicted response values.

Details

This function covers a broad range of data, Euclidean and spherical, along with their combinations.

See Also

knnreg.tune, spher.reg, spml.reg

Examples

# NOT RUN {
y <- iris[, 1]
x <- as.matrix(iris[, 2:4])
x <- x/ sqrt( rowSums(x^2) )  ## Euclidean response and spherical predictors
a <- knn.reg(x, y, x, k = 5, res = "eucl", type = "spher", estim = "arithmetic")

y <- iris[, 2:4]
y <- y / sqrt( rowSums(y^2) )  ## Spherical response and Euclidean predictor
x <- iris[, 1]
a <- knn.reg(x, y, x, k = 5, res = "spher", type = "euclidean", estim = "arithmetic")
# }