Add AIC method for mmkin column objects

Static documentation rebuilt by pkgdown

2 files changed, 52 insertions, 4 deletions

diff --git a/man/AIC.mmkin.Rd b/man/AIC.mmkin.Rd
new file mode 100644
index 00000000..e7f5c228
--- /dev/null
+++ b/man/AIC.mmkin.Rd
@@ -0,0 +1,44 @@
+\name{AIC.mmkin}
+\alias{AIC.mmkin}
+\title{
+  Calculated the AIC for a column of an mmkin object
+}
+\description{
+  Provides a convenient way to compare different kineti models fitted to the
+  same dataset.
+}
+\usage{
+  \method{AIC}{mmkin}(object, ..., k = 2)
+}
+\arguments{
+  \item{object}{
+    An object of class \code{\link{mmkin}}, containing only one column.
+  }
+  \item{\dots}{
+    For compatibility with the generic method
+  }
+  \item{k}{
+    As in the generic method
+  }
+}
+\value{
+  As in the generic method (a numeric value for single fits, or a dataframe if
+  there are several fits in the column).
+}
+\examples{
+  f <- mmkin(c("SFO", "FOMC", "DFOP"),
+    list("FOCUS A" = FOCUS_2006_A,
+         "FOCUS C" = FOCUS_2006_C))
+  AIC(f[1, "FOCUS A"]) # We get a single number for a single fit
+
+  # For FOCUS A, the models fit almost equally well, so the higher the number
+  # of parameters, the higher (worse) the AIC
+  AIC(f[, "FOCUS A"])
+  AIC(f[, "FOCUS A"], k = 0) # If we do not penalize additional parameters, we get nearly the same
+
+  # For FOCUS C, the more complex models fit better
+  AIC(f[, "FOCUS C"])
+}
+\author{
+  Johannes Ranke
+}
diff --git a/man/logLik.mkinfit.Rd b/man/logLik.mkinfit.Rd
index 8080f3db..fe517955 100644
--- a/man/logLik.mkinfit.Rd
+++ b/man/logLik.mkinfit.Rd
@@ -7,8 +7,8 @@
   This function simply calculates the product of the likelihood densities
   calculated using \code{\link{dnorm}}, i.e. assuming normal distribution.
 
-  The total number of estimated parameters returned with the value 
-  of the likelihood is calculated as the sum of fitted degradation 
+  The total number of estimated parameters returned with the value
+  of the likelihood is calculated as the sum of fitted degradation
   model parameters and the fitted error model parameters.
 
   For the case of unweighted least squares fitting, we calculate one
@@ -17,7 +17,7 @@
 
   For the case of manual weighting, we use the weight given for each
   observation as standard deviation in calculating its likelihood
-  and the total number of estimated parameters is equal to the 
+  and the total number of estimated parameters is equal to the
   number of fitted degradation model parameters.
 
   In the case of iterative reweighting, the variances obtained by this
@@ -28,7 +28,7 @@
   reweighting method is "tc".
 }
 \usage{
-\method{logLik}{mkinfit}(object, ...)
+  \method{logLik}{mkinfit}(object, ...)
 }
 \arguments{
   \item{object}{
@@ -43,6 +43,10 @@
   estimated parameters (degradation model parameters plus variance
   model parameters) as attribute.
 }
+\seealso{
+  Compare the AIC of columns of \code{\link{mmkin}} objects using
+  \code{\link{AIC.mmkin}}.
+}
 \examples{
   sfo_sfo <- mkinmod(
     parent = mkinsub("SFO", to = "m1"),