Restrictions for KZS Parameters

Share:

Description

For a d-dimensional vector of input variables, this function will calculate the values by which the parameters smooth and scale are bounded by.

Usage

1
kzs.params(x, dimension)

Arguments

x

a matrix or data frame containing the input variable(s) that is to be used in kzs, kzs.2d, or kzs.md.

dimension

an integer specifying the dimensionality of x; i.e, the number of columns in x.

Details

The compilation of functions within the kzs package requires the specification of two parameters: the first is smooth, the range of smoothing along each variable in x; the second is scale, a scale reading of each corresponding input variable in x. Each parameter is subject to two restrictions; smooth[i] and scale[i] must be positive real numbers; scale[i] must be less than or equal to the difference of sorted, consecutive x[,i] values and smooth[i] must be much less than the difference of the maximum and minimum values for its corresponding input variable, x[,i]. For each input variable in x, there must be a corresponding smooth and scale. This function was developed to be used prior to any of the functions within kzs in order to increase the efficiency of use.

Author(s)

Derek Cyr cyr.derek@gmail.com and Igor Zurbenko igorg.zurbenko@gmail.com

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
# Generate 3 random sequences of numbers that would act as the input data set 
x1 <- rnorm(100, 3, 6)
x2 <- rnorm(100, 4, 5)
x3 <- runif(100, 0, 1)

# A matrix or a data frame will work 
mat <- matrix(c(x1, x2, x3), nrow = 100, ncol = 3)

# Dimensionality is 3 since there are 3 input variables
kzs.params(x = mat, dimension = 3)