Generalised inner product and methods for class "MixComp"

Generalised inner product and methods for class MixComp. The methods for MixComp provide for very convenient computing with MixAR models.

Usage

inner(x, y, star = "*", plus = .mplus)

Arguments

x: the first argument.
y: the second argument.
star: function to apply to pairs of elements from x and y, default is multiplication, as for the usual inner product.
plus: function to apply to combine the results from the pairs, default is addition, as for the usual inner product.

Details

inner computes a generalised inner product x . y, where multiplication and summation can be replaced by other functions.

The default method of inner applies star to the corresponding pairs of elements and combines them with plus. There is no recycling, if x and y have different lengths, an error is raised. The elements of x and y are accessed with "[[". plus should be an n-ary operation.

Value

the inner product, the type of the result depends on the arguments

Methods

Methods for inner product between a "MixComp" object and a vector are similar to a product between a matrix and a vector but comply with the conventions of class "MixComp". For this reason they are described in the help page for class "MixComp", along with methods for other functions and operators applied to "MixComp" objects.

signature(x = "ANY", y = "ANY", star = "ANY", plus = "ANY"): This is the default method, see section Details.
signature(x = "MixComp", y = "missing", star = "missing", plus = "missing"): see "MixComp".
signature(x = "MixComp", y = "numeric", star = "missing", plus = "missing"): see "MixComp".
signature(x = "numeric", y = "MixComp", star = "missing", plus = "missing"): see "MixComp".
signature(x = "MixComp", y = "numeric", star = "ANY", plus = "ANY"): see "MixComp".
signature(x = "MixComp", y = "numeric", star = "ANY", plus = "missing"): see "MixComp".

Examples

inner(1:3, 2:4) # [1] 20
#> [1] 20
class(inner(1:3, 2:4)) # [1] "integer"
#> [1] "integer"
## compare to:
1:3 %*% 2:4        # 20, but (1,1)-matrix
#>      [,1]
#> [1,]   20
class(1:3 %*% 2:4) # matrix
#> [1] "matrix" "array"