From ad0efc2d16a84c674307ad2df9d44153b44a9cf8 Mon Sep 17 00:00:00 2001 From: Johannes Ranke Date: Thu, 20 Apr 2023 19:51:37 +0200 Subject: Small changes to the multistart vignette - Fix legend positions and y axis scaling - Include two more variants in the model comparison at the end, and fix the logic of the discussion --- vignettes/web_only/multistart.rmd | 27 +++++++++++---------------- 1 file changed, 11 insertions(+), 16 deletions(-) diff --git a/vignettes/web_only/multistart.rmd b/vignettes/web_only/multistart.rmd index 27a8a96a..2a9b3599 100644 --- a/vignettes/web_only/multistart.rmd +++ b/vignettes/web_only/multistart.rmd @@ -1,7 +1,7 @@ --- title: Short demo of the multistart method author: Johannes Ranke -date: Last change 26 September 2022 (rebuilt `r Sys.Date()`) +date: Last change 20 April 2023 (rebuilt `r Sys.Date()`) output: html_document vignette: > @@ -38,9 +38,9 @@ function tells us that the confidence interval for the standard deviation of 'log_k2' includes zero. We check this assessment using multiple runs with different starting values. -```{r} +```{r, warnings = FALSE} f_saem_full_multi <- multistart(f_saem_full, n = 16, cores = 16) -parplot(f_saem_full_multi) +parplot(f_saem_full_multi, lpos = "topleft") ``` This confirms that the variance of k2 is the most problematic parameter, so we @@ -51,7 +51,7 @@ for k2. f_saem_reduced <- update(f_saem_full, no_random_effect = "log_k2") illparms(f_saem_reduced) f_saem_reduced_multi <- multistart(f_saem_reduced, n = 16, cores = 16) -parplot(f_saem_reduced_multi, lpos = "topright") +parplot(f_saem_reduced_multi, lpos = "topright", ylim = c(0.5, 2)) ``` The results confirm that all remaining parameters can be determined with sufficient @@ -63,20 +63,15 @@ We can also analyse the log-likelihoods obtained in the multiple runs: llhist(f_saem_reduced_multi) ``` -The parameter histograms can be further improved by excluding the result with -the low likelihood. - -```{r} -parplot(f_saem_reduced_multi, lpos = "topright", llmin = -326, ylim = c(0.5, 2)) -``` - -We can use the `anova` method to compare the models, including a likelihood ratio -test if the models are nested. +We can use the `anova` method to compare the models. ```{r} -anova(f_saem_full, best(f_saem_reduced_multi), test = TRUE) +anova(f_saem_full, best(f_saem_full_multi), + f_saem_reduced, best(f_saem_reduced_multi)) ``` -While AIC and BIC are lower for the reduced model, the likelihood ratio test -does not indicate a significant difference between the fits. +The reduced model gives the lowest information criteria and similar +likelihoods as the best variant of the full model. The multistart method +leads to a much lower improvement of the likelihood for the reduced model, +indicating that it converges faster. -- cgit v1.2.1