confint.BranchGLM {BranchGLM}

Likelihood Ratio Confidence Intervals for Beta Coefficients for BranchGLM Objects

Description

Finds profile likelihood ratio confidence intervals for beta coefficients with the ability to calculate the intervals in parallel.

Usage

## S3 method for class 'BranchGLM'
confint(object, parm, level = 0.95, parallel = FALSE, nthreads = 8, ...)

Arguments

object

a BranchGLM object.

parm

a specification of which parameters are to be given confidence intervals, either a vector of numbers or a vector of names. If missing, all parameters are considered.

level

the confidence level required.

parallel

a logical value to indicate if parallelization should be used.

nthreads

a positive integer to denote the number of threads used with OpenMP, only used if parallel = TRUE.

...

further arguments passed from other methods.

Details

Endpoints of the confidence intervals that couldn't be found by the algorithm are filled in with NA. When there is a lot of multicollinearity in the data the algorithm may have problems finding many of the intervals.

Value

An object of class BranchGLMCIs which is a list with the following components.

CIs

a numeric matrix with the confidence intervals

level

the supplied level

MLE

a numeric vector of the MLEs of the coefficients

See Also

plot.BranchGLMCIs, plotCI

Examples

Data <- iris
### Fitting linear regression model
mymodel <- BranchGLM(Sepal.Length ~ ., data = Data, family = "gaussian", link = "identity")

### Getting confidence intervals
CIs <- confint(mymodel, level = 0.95)
CIs

### Plotting CIs
plot(CIs, mary = 7, cex.y = 0.9)