Hypothesis-Testing and Related Methods for "nestedLogit" Objects

Various methods for testing hypotheses about nested logit models.

Anova: Calculates type-II or type-III analysis-of-variance tables for "nestedLogit" objects; see Anova in the car package.
anova: Computes sequential analysis of variance (or deviance) tables for one or more fitted "nestedLogit" objects; see anova.
linearHypothesis: Computes Wald tests for linear hypotheses; see linearHypothesis in the car package.
logLik: Returns the log-likelihood and degrees of freedom for the nested-dichotomies model. (and through it AIC and BIC model-comparison statistics).

Usage

# S3 method for nestedLogit
Anova(mod, ...)

# S3 method for Anova.nestedLogit
print(x, ...)

# S3 method for nestedLogit
linearHypothesis(model, ...)

# S3 method for nestedLogit
anova(object, object2, ...)

# S3 method for anova.nestedLogit
print(x, ...)

# S3 method for nestedLogit
logLik(object, ...)

Arguments

...: arguments to be passed down. In the case of linearHypothesis, the second argument is typically the hypothesis.matrix. See the Details section of linearHypothesis. In the case of anova, additional sequential "nestedLogit" models.
x, object, object2, mod, model: in most cases, an object of class "nestedLogit".

Value

The Anova and anova methods return objects of class "Anova.nestedLogit" and "anova.nestedLogit", respectively, each of which contains a list of "anova" objects (see anova) and is usually printed.
The linearHypothesis method is called for its side effect, printing the result of linear hypothesis tests, and invisibly returns NULL.
The logLik method returns an object of class "logLik" (see logLik).

Author

John Fox

Examples

# define continuation dichotomies for level of education
cont.dichots <- continuationLogits(c("l.t.highschool",
                                     "highschool",
                                     "college",
                                     "graduate"))
# fit a nested model for the GSS data examining education degree in relation to parent & year
m <- nestedLogit(degree ~ parentdeg + year,
                 cont.dichots,
                 data=GSS)

# Anova and anova tests
car::Anova(m) # type-II (partial) tests
#> 
#>  Analysis of Deviance Tables (Type II tests)
#>  
#> Response above_l.t.highschool: {l.t.highschool} vs. {highschool, college, graduate}
#>           LR Chisq Df Pr(>Chisq)    
#> parentdeg   6604.2  3  < 2.2e-16 ***
#> year         383.3  1  < 2.2e-16 ***
#> ---
#> Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#> 
#> 
#> Response above_highschool: {highschool} vs. {college, graduate}
#>           LR Chisq Df Pr(>Chisq)    
#> parentdeg   3541.7  3  < 2.2e-16 ***
#> year         159.8  1  < 2.2e-16 ***
#> ---
#> Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#> 
#> 
#> Response above_college: {college} vs. {graduate}
#>           LR Chisq Df Pr(>Chisq)    
#> parentdeg  121.317  3  < 2.2e-16 ***
#> year        29.074  1  6.966e-08 ***
#> ---
#> Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#> 
#> 
#> Combined Responses
#>           LR Chisq Df Pr(>Chisq)    
#> parentdeg  10267.2  9  < 2.2e-16 ***
#> year         572.1  3  < 2.2e-16 ***
#> ---
#> Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

anova(update(m, . ~ . - year), m) # model comparison
#> 
#>  Analysis of Deviance Tables
#>  Model 1: degree ~ parentdeg
#>  Model 2: degree ~ parentdeg + year 
#>  
#> Response above_l.t.highschool: {l.t.highschool} vs. {highschool, college, graduate}
#>   Resid. Df Resid. Dev Df Deviance  Pr(>Chi)    
#> 1     44087      33263                          
#> 2     44086      32880  1   383.29 < 2.2e-16 ***
#> ---
#> Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#> 
#> 
#> Response above_highschool: {highschool} vs. {college, graduate}
#>   Resid. Df Resid. Dev Df Deviance  Pr(>Chi)    
#> 1     36340      40853                          
#> 2     36339      40693  1   159.75 < 2.2e-16 ***
#> ---
#> Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#> 
#> 
#> Response above_college: {college} vs. {graduate}
#>   Resid. Df Resid. Dev Df Deviance  Pr(>Chi)    
#> 1     11095      14074                          
#> 2     11094      14045  1   29.074 6.966e-08 ***
#> ---
#> Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#> 
#> 
#> Combined Responses
#>   Resid. Df Resid. Dev Df Deviance  Pr(>Chi)    
#> 1     91522      88190                          
#> 2     91519      87618  3   572.11 < 2.2e-16 ***
#> ---
#> Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

# Wald test
car::linearHypothesis(m, c("parentdeghighschool", "parentdegcollege",
                           "parentdeggraduate"))
#> Linear hypothesis test
#> 
#> Hypothesis: 
#> parentdeghighschool = 0 
#> parentdegcollege = 0 
#> parentdeggraduate = 0 
#>  
#> Model 1: restricted model
#> Model 2: degree ~ parentdeg + year 
#>  
#> Response above_l.t.highschool: {l.t.highschool} vs. {highschool, college, graduate} 
#>   Res.Df Df Chisq Pr(>Chisq)    
#> 1  44089                        
#> 2  44086  3  5085  < 2.2e-16 ***
#> ---
#> Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#> 
#> Response above_highschool: {highschool} vs. {college, graduate}
#>   Res.Df Df  Chisq Pr(>Chisq)    
#> 1  36342                         
#> 2  36339  3 3322.4  < 2.2e-16 ***
#> ---
#> Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#> 
#> Response above_college: {college} vs. {graduate}
#>   Res.Df Df  Chisq Pr(>Chisq)    
#> 1  11097                         
#> 2  11094  3 122.17  < 2.2e-16 ***
#> ---
#> Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#> 
#> Combined Responses
#> Chisq = 8529.624, Df = 9, Pr(>Chisq) = < 2.22e-16

# log-liklihood, AIC, and BIC
logLik(m)
#> 'log Lik.' -43809.17 (df=15)
AIC(m)
#> [1] 87648.33
BIC(m)
#> [1] 87778.74

Hypothesis-Testing and Related Methods for `"nestedLogit"` Objects

Usage

Arguments

Value

See also

Author

Examples