Relabel the Components

The components are sorted by the value of one of the parameters or according to an integer vector containing the permutation of the numbers from 1 to the number of components.

relabel(object, by, ...)
# S4 method for class 'FLXdist,character'
relabel(object, by, which = NULL, ...)

Arguments

object

An object of class "flexmix".

by

If a character vector, it needs to be one of "prior", "model", "concomitant" indicating if the parameter should be from the component-specific or the concomitant variable model. If an integer vector it indicates how the components should be sorted. If missing, the components are sorted by component size.

which

Name (or unique substring) of a parameter if by is equal to "model" or "concomitant".

...

Currently not used.

Author

Friedrich Leisch and Bettina Gruen

Examples

    set.seed(123)
    beta <- matrix(1:16, ncol = 4)
    beta
#>      [,1] [,2] [,3] [,4]
#> [1,]    1    5    9   13
#> [2,]    2    6   10   14
#> [3,]    3    7   11   15
#> [4,]    4    8   12   16
    df1 <- ExLinear(beta, n = 100, sd = .5)
    f1 <- flexmix(y~., data = df1, k = 4)

    ## There was label switching, parameters are not in the same order
    ## as in beta:
    round(parameters(f1))
#>                  Comp.1 Comp.2 Comp.3 Comp.4
#> coef.(Intercept)      1      5      9     13
#> coef.x1               2      6     10     14
#> coef.x2               3      7     11     15
#> coef.x3               4      8     12     16
#> sigma                 0      0      1      1
        
    betas <- rbind(beta, .5)
    betas
#>      [,1] [,2] [,3] [,4]
#> [1,]  1.0  5.0  9.0 13.0
#> [2,]  2.0  6.0 10.0 14.0
#> [3,]  3.0  7.0 11.0 15.0
#> [4,]  4.0  8.0 12.0 16.0
#> [5,]  0.5  0.5  0.5  0.5

    ## This makes no sense:
    summary(abs(as.vector(betas-parameters(f1))))
#>     Min.  1st Qu.   Median     Mean  3rd Qu.     Max. 
#> 0.007918 0.025205 0.063959 0.092210 0.117254 0.347028 

    ## We relabel the components by sorting the coefficients of x1:
    r1 <- relabel(f1, by = "model", which = "x1")
    round(parameters(r1))
#>                  Comp.1 Comp.2 Comp.3 Comp.4
#> coef.(Intercept)      1      5      9     13
#> coef.x1               2      6     10     14
#> coef.x2               3      7     11     15
#> coef.x3               4      8     12     16
#> sigma                 0      0      1      1

    ## Now we can easily compare the fit with the true parameters:
    summary(abs(as.vector(betas-parameters(r1))))
#>     Min.  1st Qu.   Median     Mean  3rd Qu.     Max. 
#> 0.007918 0.025205 0.063959 0.092210 0.117254 0.347028