dr4pl (version 1.0.0)

confint.dr4pl: Fit a 4 parameter logistic (4PL) model to dose-response data.

Description

Compute the approximate confidence intervals of the parameters of a 4PL model based on the asymptotic normality of least squares estimators.

Usage

# S3 method for dr4pl
confint(object, parm = NULL, level = 0.95, ...)

Arguments

object

An object of the dr4pl class

parm

Parameters of the 4PL model

level

Confidence level

...

Other parameters to be passed

Value

A matrix of the confidence intervals in which each row represents a parameter and each column represents the lower and upper bounds of the confidence intervals of the corresponding parameters.

Details

This function computes the approximate confidence intervals of the true parameters of a 4PL model based on the asymptotic normality of the least squares estimators in nonlinear regression. The Hessian matrix is used to obtain the second order approximation to the sum-of-squares loss function. Please refer to Subsection 5.2.2 of Seber and Wild (1989).

References

Seber1989dr4pl

Examples

# NOT RUN {
obj.dr4pl <- dr4pl(Response ~ Dose, data = sample_data_1)  # Fit a 4PL model to data

confint(obj.dr4pl)  # Print conventional 95% confidence intervals
confint(obj.dr4pl, level = 0.99)  # Print 99%confidence intervals

theta <- coef(obj.dr4pl)
theta[4] <- 0  # Set the lower asymptote to be zero
confint(obj.dr4pl, parm = theta)  # Use our dr4pl object but different parameter estimates

# }