df_term: Calculate degree of freedom of a modelterm (contrast) for a...

In predictmeans: Predicted Means for Linear and Semiparametric Models

Calculate degree of freedom of a modelterm (contrast) for a lmer model

Description

Calculate the degree of freedom of a modelterm (contrast) for a lmer model using "Kenward-Roger" or "Satterthwaite" method.

Usage

  df_term(model, modelterm, ctrmatrix=NULL, ctrnames=NULL, 
          type=c("Kenward-Roger", "Satterthwaite"))

Arguments

model	Model object returned by lmer.
modelterm	Name (in "quotes") for indicating which factor term's degree of freedom to be calculated. The modelterm must be given exactly as it appears in the model formlar, e.g. "A" or "A:B".
ctrmatrix	A specified contrast matrix. If ctrmatrix isn't NULL, the programe will ignore modelterm and calculate degree of freedom for the ctrmatrix.
ctrnames	Names of the specified contrasts, e.g. c("A vs D", "C vs B", ...)
type	Name (in "quote") for indicating a method for claculating degree of freedom. The choices are "Kenward-Roger" and "Satterthwaite". The default method is "Kenward-Roger".

Author(s)

Dongwen Luo, Siva Ganesh and John Koolaard

Examples

library(predictmeans)
# ftable(xtabs(yield ~ Block+Variety+nitro, data=Oats))
Oats$nitro <- factor(Oats$nitro)
fm <- lmer(yield ~ nitro*Variety+(1|Block/Variety), data=Oats)
df_term(fm, "nitro:Variety")
## Not run: 
## The contrast has a contrast matrix as follows:
#     0:Golden Rain 0:Marvellous 0:Victory 
#[1,]            -1            0         1 
#[2,]             0            0         1 
#     0.2:Golden Rain 0.2:Marvellous 0.2:Victory 
#[1,]               0              0           0 
#[2,]               0              0           0 
#     0.4:Golden Rain  0.4:Marvellous 0.4:Victory
#[1,]               0               0           0
#[2,]               0              -1           0
#      0.6:Golden Rain 0.6:Marvellous 0.6:Victory
#[1,]                0              0           0
#[2,]                0              0           0

# 1. Enter above contrast matrix into a pop up window, then close the window
# df_term(fm, "nitro:Variety")
 
# 2. Construct the contrast matrix directly
cm <- rbind(c(-1, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0), 
            c(0, 0, 1, 0, 0, 0, 0, -1, 0, 0, 0, 0))
df_term(fm, ctrmatrix=cm, type="Satterthwaite")

predictmeans documentation built on Oct. 20, 2023, 5:07 p.m.