From aa5aeeb5d55017774eff1066a495df59b97ca233 Mon Sep 17 00:00:00 2001
From: Johannes Ranke ‘plot_res’, ‘plot_sep’ and ‘mkinerrplot’: Add the possibility to show standardized residuals and make it the default for fits with error models other than ‘const’ ‘lrtest.mkinfit’: Improve naming of the compared fits in the case of fixed parameters ‘confint.mkinfit’: Make the quadratic approximation the default, as the likelihood profiling takes a lot of time, especially if the fit has more than three parameters
+
diff --git a/docs/reference/mkinresplot.html b/docs/reference/mkinresplot.html
index 1ac2cdbe..4429798b 100644
--- a/docs/reference/mkinresplot.html
+++ b/docs/reference/mkinresplot.html
@@ -72,7 +72,7 @@ argument show_residuals = TRUE." />
@@ -143,7 +143,8 @@ argument show_residuals = TRUE
.
mkinresplot(object, obs_vars = names(object$mkinmod$map), xlim = c(0, - 1.1 * max(object$data$time)), xlab = "Time", ylab = "Residual", + 1.1 * max(object$data$time)), standardized = FALSE, xlab = "Time", + ylab = ifelse(standardized, "Standardized residual", "Residual"), maxabs = "auto", legend = TRUE, lpos = "topright", col_obs = "auto", pch_obs = "auto", frame = TRUE, ...)@@ -164,14 +165,18 @@ the model
plot range in x direction.
Should the residuals be standardized by dividing by the +standard deviation given by the error model of the fit?
Label for the x axis. Defaults to "Time [days]".
Label for the x axis.
Label for the y axis. Defaults to "Residual [% of applied -radioactivity]".
Label for the y axis.
Should a legend be plotted? Defaults to "TRUE".
Should a legend be plotted?
legendSee also
mkinplot
, for a way to plot the data and the fitted
- lines of the mkinfit object.
+ lines of the mkinfit object, and plot_res
for a function
+ combining the plot of the fit and the residual plot.
Examples
diff --git a/docs/reference/plot.mkinfit-2.png b/docs/reference/plot.mkinfit-2.png
index 6910ae47..72bc331e 100644
Binary files a/docs/reference/plot.mkinfit-2.png and b/docs/reference/plot.mkinfit-2.png differ
diff --git a/docs/reference/plot.mkinfit-3.png b/docs/reference/plot.mkinfit-3.png
index 8d399598..72bc331e 100644
Binary files a/docs/reference/plot.mkinfit-3.png and b/docs/reference/plot.mkinfit-3.png differ
diff --git a/docs/reference/plot.mkinfit-4.png b/docs/reference/plot.mkinfit-4.png
index 20f30221..8d399598 100644
Binary files a/docs/reference/plot.mkinfit-4.png and b/docs/reference/plot.mkinfit-4.png differ
diff --git a/docs/reference/plot.mkinfit-5.png b/docs/reference/plot.mkinfit-5.png
index ad2ffa8c..20f30221 100644
Binary files a/docs/reference/plot.mkinfit-5.png and b/docs/reference/plot.mkinfit-5.png differ
diff --git a/docs/reference/plot.mkinfit-6.png b/docs/reference/plot.mkinfit-6.png
new file mode 100644
index 00000000..0b3769ce
Binary files /dev/null and b/docs/reference/plot.mkinfit-6.png differ
diff --git a/docs/reference/plot.mkinfit-7.png b/docs/reference/plot.mkinfit-7.png
new file mode 100644
index 00000000..ad2ffa8c
Binary files /dev/null and b/docs/reference/plot.mkinfit-7.png differ
diff --git a/docs/reference/plot.mkinfit.html b/docs/reference/plot.mkinfit.html
index 04ef75e4..a3b912bb 100644
--- a/docs/reference/plot.mkinfit.html
+++ b/docs/reference/plot.mkinfit.html
@@ -71,7 +71,7 @@ observed data together with the solution of the fitted model." />
@@ -150,9 +150,13 @@ observed data together with the solution of the fitted model.
lpos = "topright", inset = c(0.05, 0.05), show_errmin = FALSE,
errmin_digits = 3, frame = TRUE, ...)
-plot_sep(fit, show_errmin = TRUE, ...)
+plot_sep(fit, show_errmin = TRUE,
+ show_residuals = ifelse(identical(fit$err_mod, "const"), TRUE,
+ "standardized"), ...)
-plot_res(fit, sep_obs = FALSE, show_errmin = sep_obs, ...)
+plot_res(fit, sep_obs = FALSE, show_errmin = sep_obs,
+ show_residuals = ifelse(identical(fit$err_mod, "const"), TRUE,
+ "standardized"), ...)
plot_err(fit, sep_obs = FALSE, show_errmin = sep_obs, ...)
@@ -216,7 +220,9 @@ corresponding model prediction lines.
Should residuals be shown? If only one plot of the
fits is shown, the residual plot is in the lower third of the plot.
Otherwise, i.e. if "sep_obs" is given, the residual plots will be located
-to the right of the plots of the fitted curves.
+to the right of the plots of the fitted curves. If this is set to
+'standardized', a plot of the residuals divided by the standard deviation
+ given by the fitted error model will be shown.
Further arguments passed to plot
.
For