Upgrade Objects to Common Class — upgrade_objects • rationalmatrices

Used in cbind, rbind, ratm_mult, and Ops.ratm to make operations on different subclasses of the (super-) class ratm possible.

Usage

upgrade_objects(force = TRUE, ...)

Arguments

force: Default set to TRUE. If FALSE, then objects are also coerced when only one object is supplied. Option FALSE is used in bind methods
...: Arbitrary number of objects with superclass ratm

Value

Upgraded inputs

Details

Constant matrices are always upgraded to polm, and lmfd and rmfd are always upgraded to stsp if two ratm objects are involved. Therefore, most operations are performed in the state space setting.

Depending on the highest "grade" of any object involved in the operation, the lower ranked object is upgraded to the highest one. The following ranks are assigned to the ratm objects:

1: polm
2: lpolm
3: lmfd
4: rmfd
5: stsp
6: pseries
7: zvalues

Laurent polynomials lpolm have a special place because they are detached from/unconnected to elements lmfd, rmfd, stsp, pseries, zvalues.

Operations with lpolm are only possible with polm objects (and those which can be coerced to polm).

Operations which are not defined are

pseries and zvalues: pseries cannot be coerced (easily) to zvalues object
Anything involving lpolm except for polm objects

Examples

p = test_polm(degree = 2)
lp = test_lpolm(degree_max = 1, degree_min = -1)
l = test_lmfd()
r = test_lmfd()
ss = test_stsp(s = 1)
ps = pseries(ss)
zv = zvalues(ss)

# Only one argument
rationalmatrices:::upgrade_objects(force = TRUE, l)
#> [[1]]
#> statespace realization [1,1] with s = 1 states
#>          s[1]     u[1]
#> s[1] 1.181889 4.578742
#> x[1] 1.000000 1.856831
#> 
rationalmatrices:::upgrade_objects(force = FALSE, l)
#> [[1]]
#> ( 1 x 1 ) left matrix fraction description a^(-1)(z) b(z) with degrees (p = 1, q = 1)
#> left factor a(z):
#>      z^0 [,1]  z^1 [,1]
#> [1,]        1 -1.181889
#> right factor b(z):
#>      z^0 [,1] z^1 [,1]
#> [1,] 1.856831 2.384173
#> 

# Upgrade poly to rmfd
rationalmatrices:::upgrade_objects(force = TRUE, p, r)
#> [[1]]
#> statespace realization [1,1] with s = 2 states
#>      s[1] s[2] u[1]
#> s[1]    0    0    1
#> s[2]    1    0    0
#> x[1]  111  112  110
#> 
#> [[2]]
#> statespace realization [1,1] with s = 1 states
#>            s[1]       u[1]
#> s[1] -0.4218751  0.1254212
#> x[1]  1.0000000 -2.0095552
#> 
rationalmatrices:::upgrade_objects(force = TRUE, l, ps)
#> [[1]]
#> ( 1 x 1 ) impulse response with maximum lag = 5 
#>      lag=0 [,1] lag=1 [,1] lag=2 [,1] lag=3 [,1] lag=4 [,1] lag=5 [,1]
#> [1,]   1.856831   4.578742   5.411566   6.395872   7.559212   8.934152
#> 
#> [[2]]
#> ( 1 x 1 ) impulse response with maximum lag = 5 
#>      lag=0 [,1] lag=1 [,1]   lag=2 [,1]    lag=3 [,1]   lag=4 [,1]
#> [1,]          1 -0.3728132 0.0009012476 -2.178698e-06 5.266837e-09
#>         lag=5 [,1]
#> [1,] -1.273218e-11
#>