Simultaneous Inference in Finite Mixtures of Regression Models

Extracts coefficient estimates and their covariance estimate to perform tests for zero coefficients or constant effects re-using functionality from package multcomp.

flxglht(model, linfct, ...)
# S4 method for class 'flexmix,character'
flxglht(model, linfct, ...)
# S4 method for class 'FLXRoptim,character'
flxglht(model, linfct, ...)

Arguments

model

Either a fitted mixture model of class "flexmix" or a re-fitted mixture model of class "FLXRoptim".

linfct

A character string which can be either "zero" or "tukey".

...

Currently not used.

Details

Only tested for finite mixture models fitted with driver FLXMRglm.

Value

An object of class "glht".

References

Friedrich Leisch and Torsten Hothorn. Simultaneous Inference in Finite Mixtures of Regression Models. Austrian Journal of Statistics, forthcoming.

Author

Friedrich Leisch

See also

glht

Examples

data("NPreg", package = "flexmix")

ex1 <- flexmix(yn ~ x + I(x^2), data = NPreg, k = 2,
               control = list(verb = 5, iter = 100))
#> Classification: weighted 
#>    5 Log-likelihood :    -720.7965 
#>   10 Log-likelihood :    -649.6249 
#>   15 Log-likelihood :    -642.5451 
#> converged

zero_effect <- flxglht(ex1, "zero")
zero_effect
#> 
#> 	 General Linear Hypotheses
#> 
#> Linear Hypotheses:
#>                     Estimate
#> C1.(Intercept) == 0 -0.20851
#> C2.(Intercept) == 0 14.71325
#> C1.x == 0            4.81446
#> C2.x == 0            9.84830
#> C1.I(x^2) == 0       0.03654
#> C2.I(x^2) == 0      -0.96852
#> 
summary(zero_effect)
#> 
#> 	 Simultaneous Tests for General Linear Hypotheses
#> 
#> Linear Hypotheses:
#>                     Estimate Std. Error z value Pr(>|z|)    
#> C1.(Intercept) == 0 -0.20851    1.00921  -0.207    0.999    
#> C2.(Intercept) == 0 14.71325    1.32377  11.115   <1e-04 ***
#> C1.x == 0            4.81446    0.50961   9.447   <1e-04 ***
#> C2.x == 0            9.84830    0.59158  16.647   <1e-04 ***
#> C1.I(x^2) == 0       0.03654    0.04976   0.734    0.865    
#> C2.I(x^2) == 0      -0.96852    0.05526 -17.526   <1e-04 ***
#> ---
#> Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
#> (Adjusted p values reported -- single-step method)
#> 

comp_effect <- flxglht(ex1, "tukey")
comp_effect
#> 
#> 	 General Linear Hypotheses
#> 
#> Linear Hypotheses:
#>                                    Estimate
#> C2.(Intercept)-C1.(Intercept) == 0   14.922
#> C2.x-C1.x == 0                        5.034
#> C2.I(x^2)-C1.I(x^2) == 0             -1.005
#> 
summary(comp_effect)
#> 
#> 	 Simultaneous Tests for General Linear Hypotheses
#> 
#> Linear Hypotheses:
#>                                    Estimate Std. Error z value Pr(>|z|)    
#> C2.(Intercept)-C1.(Intercept) == 0 14.92176    1.59810   9.337   <1e-10 ***
#> C2.x-C1.x == 0                      5.03383    0.74393   6.766   <1e-10 ***
#> C2.I(x^2)-C1.I(x^2) == 0           -1.00506    0.07098 -14.160   <1e-10 ***
#> ---
#> Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
#> (Adjusted p values reported -- single-step method)
#>