Skip to contents

Example 2.B.4 from Generalized Linear Mixed Models: Modern Concepts, Methods and Applications by Walter W. Stroup(p-56)

Source: R/Exam2.B.4.R
Exam2.B.4.Rd

Exam2.B.4 is used to illustrate one way treatment design with Binomial observations.

References

  1. Stroup, W. W. (2012). Generalized Linear Mixed Models: Modern Concepts, Methods and Applications. CRC Press.

See also

DataExam2.B.4

Author

  1. Muhammad Yaseen (myaseen208@gmail.com)

  2. Adeela Munawar (adeela.uaf@gmail.com)

Examples

#-----------------------------------------------------------------------------------
## logit Model  discussed in Example 2.B.2 using DataExam2.B.4
## Default link is logit
## using fmaily=binomial gives warning message of no-integer successes
#-----------------------------------------------------------------------------------
data(DataExam2.B.4)
DataExam2.B.4$trt <- factor(x =  DataExam2.B.4$trt)
Exam2.B.4glm <-
                glm(
                      formula = Yij/Nij ~ trt
                    , family  =  quasibinomial(link = "probit")
                    , data    = DataExam2.B.4
                    )
summary(Exam2.B.4glm)
#> 
#> Call:
#> glm(formula = Yij/Nij ~ trt, family = quasibinomial(link = "probit"), 
#>     data = DataExam2.B.4)
#> 
#> Coefficients:
#>             Estimate Std. Error t value Pr(>|t|)
#> (Intercept)  -0.7729     0.3737  -2.068    0.130
#> trt2          0.7729     0.5017   1.541    0.221
#> trt3          0.9408     0.5029   1.871    0.158
#> 
#> (Dispersion parameter for quasibinomial family taken to be 0.1426701)
#> 
#>     Null deviance: 1.01671  on 5  degrees of freedom
#> Residual deviance: 0.43701  on 3  degrees of freedom
#> AIC: NA
#> 
#> Number of Fisher Scoring iterations: 4
#> 
library(parameters)
model_parameters(Exam2.B.4glm)
#> Parameter   | Log-Risk |   SE |         95% CI |  t(3) |     p
#> --------------------------------------------------------------
#> (Intercept) |    -0.77 | 0.37 | [-1.55, -0.07] | -2.07 | 0.039
#> trt [2]     |     0.77 | 0.50 | [-0.19,  1.78] |  1.54 | 0.123
#> trt [3]     |     0.94 | 0.50 | [-0.03,  1.95] |  1.87 | 0.061
#> 
#> Uncertainty intervals (profile-likelihood) and p-values (two-tailed)
#>   computed using a Wald t-distribution approximation.