Class "raggedCoefS" — ragged list

Ragged list used to hold coefficients of MixAR models with seasonal AR parameters.

Objects from the Class

Objects are created by calls of the form

new("raggedCoefS", a = list(v1, v2 , ...), as = list(vs1, vs2, ...), s).

If orders p and ps are specified, a consistency check is made.

Slots

a

Object of class "list" containing AR values. Each element of the list must be "numeric"

p

Object of class "numeric" containing the lengths of components in a. If missing, it is generated based on lengths of elements of a.

as

Object of class "list" containing seasonal AR values. Each element of the list must be "numeric"

ps

Object of class "numeric" containing the lengths of elements of as. If missing, it is generated based on lengths of elements of as.

s

A single element "numeric" vector determining the seasonality in the model(monthly, quarterly, etc..).

Methods

Indexing with "[" treats a raggedCoef object as a matrix (one row for each component), while "[[" treats the object as list (it works on slot a). Specifically, "[[1]]" picks the systematic AR parameters, "[[2]]" picks seasonal AR parameters.

The replacement variants of "[" and "[[" do not change the structure of the object.

Replacement methods only work for subsets x[[i]], x[[i]][[j]], x[[i]][[j]][k] for suitable i, j and k.

i must be equal to 1 for x@a and 2 for x@as.

[

signature(x = "raggedCoefS", i = "missing", j = "missing"):

returns the complete matrix of coefficients, one row corresponding to one component, with '0's to match different orders

[

signature(x = "raggedCoefS", i = "missing", j = "missing"):

[

signature(x = "raggedCoefS", i = "numeric", j = "missing"):

[

signature(x = "raggedCoefS", i = "numeric", j = "numeric"): Indexing with "[" treats a raggedCoef object as a matrix with one row for each component and number of columns equal to max(p) + max(ps) in increasing lag. However, x[2] is equivalent to x[2,] which is different from the treatment of matrix objects in base R.

[[

signature(x = "raggedCoefS"), i = "numeric":

if i=1 selects the list of systematic AR parameters; if i=2 selects the list of seasonal AR parameters.

[[

signature(x = "raggedCoefS"), i = "numeric", j = "numeric":

[[

signature(x = "raggedCoefS"), i = "numeric", j = "numeric", k = "numeric":

j and k are used to select specific elements from the listt of interest.

Author

Davide Ravagli

See also

class "raggedCoef"

Examples

showClass("raggedCoefS")
#> Class "raggedCoefS" [package "mixAR"]
#> 
#> Slots:
#>                                               
#> Name:       as      ps       s       a       p
#> Class:    list numeric numeric    list numeric
#> 
#> Extends: "raggedCoef"

ragA <- new("raggedCoefS", a = list( c(0.5, -0.5), 1),
            as = list(0, c(0.3, -0.1) ), s = 12)
ragB <- new("raggedCoefS", a = list( c(0.5, -0.5), 1), p = c(2, 1),
            as = list(0, c(0.3, -0.1) ), ps = c(1, 2), s = 12)  # same

## Elements selection examples

ragA[]             ## matrix of coefficients
#>      [,1] [,2] [,3] [,4]
#> [1,]  0.5 -0.5  0.0  0.0
#> [2,]  1.0  0.0  0.3 -0.1
ragA[1]; ragA[1, ] ## vector of coefficients from first component
#> [1]  0.5 -0.5  0.0  0.0
#> [1]  0.5 -0.5  0.0  0.0
ragA[[2]]          ## list of seasonal AR parameters
#> [[1]]
#> [1] 0
#> 
#> [[2]]
#> [1]  0.3 -0.1
#> 
ragA[[2]][[1]]      ## vector of seasonal AR parameters from first component
#> [1] 0

## Replacement of values in 'raggedCoefS' objects


ragB[[2]] <- list(1, c(-0.5,0.5))
ragB[[2]][[2]] <- c(20, 22)
ragB[[1]][[1]][1] <- 0