ciFun: Calculates and plots confidence intervals for random effects.

In p-schaefer/psUtilities: Utility functions for ecological analysis

Description

Calculates and plots confidence intervals for random effects of lme4 models using REsim

Usage

ciFun(model, model.variable, data = NULL, col.variable = NULL, plot = T,
  stat = "median", level = 0.95, oddsRatio = F, nsim = 1000,
  sig.only = FALSE, group.variable, ...)

Arguments

model	A merMod object from lme4.
model.variable	Character. The name of the random effect variable to be computed.
data	An optional data frame used in model
col.variable	An optional variable name used to colour the plotted values
plot	Logical. Should a plot be produced?
stat	Character. median or mean
level	Numeric. Confidence interval to return
oddsRatio	Logical. should paramters be converted to odds ratios?
nsim	Numeric.Number of simulations to use
sig.only	Logical. Should only significant random effect levels be plotted?
group.variable	Character. Grouping variable for plots
...	other arguments passed to REsim.

Details

Use the Gelman sim technique to build empirical Bayes estimates. Uses the sim function in the arm package

Value

A data frame and plot of confidence intervals.

Examples

require(lme4)
m1 <- lmer(Reaction ~ 1 + (Days | Subject), sleepstudy)
ci1 <- ciFun(model=m1, model.variable= "Days")
d1<-cbpp
d1$period<-as.numeric(as.character(cbpp$period))
d1$response<-d1$incidence/d1$size
gm1 <- glmer(response ~ 1 + (period | herd), data = d1, weights=d1$size, family = binomial)
ci2 <- ciFun(model=gm1, model.variable= "period")

p-schaefer/psUtilities documentation built on May 24, 2019, 5:56 p.m.