parms and confint methods

The confint method can do profile likelihood based confidence intervals!

2 files changed, 83 insertions, 0 deletions

diff --git a/man/confint.mkinfit.Rd b/man/confint.mkinfit.Rd
new file mode 100644
index 00000000..94f55d14
--- /dev/null
+++ b/man/confint.mkinfit.Rd
@@ -0,0 +1,56 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/confint.mkinfit.R
+\name{confint.mkinfit}
+\alias{confint.mkinfit}
+\title{Confidence intervals for parameters of mkinfit objects}
+\usage{
+\method{confint}{mkinfit}(object, parm, level = 0.95, alpha = 1 -
+  level, method = c("profile", "quadratic"), transformed = TRUE,
+  backtransform = TRUE, distribution = c("student_t", "normal"),
+  quiet = FALSE, ...)
+}
+\arguments{
+\item{object}{An \code{\link{mkinfit}} object}
+
+\item{parm}{A vector of names of the parameters which are to be given
+confidence intervals. If missing, all parameters are considered.}
+
+\item{level}{The confidence level required}
+
+\item{alpha}{The allowed error probability, overrides 'level' if specified.}
+
+\item{method}{The 'profile' method searches the parameter space for the
+cutoff of the confidence intervals by means of a likelihood ratio test.
+The 'quadratic' method approximates the likelihood function at the
+optimised parameters using the second term of the Taylor expansion, using
+a second derivative (hessian) contained in the object.}
+
+\item{transformed}{If the quadratic approximation is used, should it be
+applied to the likelihood based on the transformed parameters?}
+
+\item{backtransform}{If we approximate the likelihood in terms of the
+transformed parameters, should we backtransform the parameters with
+their confidence intervals?}
+
+\item{distribution}{For the quadratic approximation, should we use
+the student t distribution or assume normal distribution for
+the parameter estimate}
+
+\item{quiet}{Should we suppress messages?}
+}
+\value{
+A matrix with columns giving lower and upper confidence limits for
+  each parameter.
+}
+\description{
+Confidence intervals for parameters of mkinfit objects
+}
+\examples{
+f <- mkinfit("SFO", FOCUS_2006_C, quiet = TRUE)
+confint(f, method = "quadratic")
+confint(f, method = "profile")
+}
+\references{
+Pawitan Y (2013) In all likelihood - Statistical modelling and
+  inference using likelihood. Clarendon Press, Oxford.
+}
diff --git a/man/parms.Rd b/man/parms.Rd
new file mode 100644
index 00000000..6de52557
--- /dev/null
+++ b/man/parms.Rd
@@ -0,0 +1,27 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/parms.mkinfit.R
+\name{parms}
+\alias{parms}
+\alias{parms.mkinfit}
+\title{Extract model parameters from mkinfit models}
+\usage{
+parms(object, ...)
+
+\method{parms}{mkinfit}(object, transformed = FALSE, ...)
+}
+\arguments{
+\item{object}{A fitted model object}
+
+\item{transformed}{Should the parameters be returned
+as used internally during the optimisation?}
+
+\item{complete}{Unused argument for compatibility with the generic coef function from base R}
+}
+\value{
+A numeric vector of fitted model parameters
+}
+\description{
+This function always returns degradation model parameters as well as error
+model parameters, in order to avoid working with a fitted model without
+considering the error structure that was assumed for the fit.
+}