diff options
Diffstat (limited to 'R/nlme.mmkin.R')
| -rw-r--r-- | R/nlme.mmkin.R | 36 |
1 files changed, 18 insertions, 18 deletions
diff --git a/R/nlme.mmkin.R b/R/nlme.mmkin.R index 82d5f6de..ff1f2fff 100644 --- a/R/nlme.mmkin.R +++ b/R/nlme.mmkin.R @@ -24,6 +24,11 @@ get_deg_func <- function() { #' This functions sets up a nonlinear mixed effects model for an mmkin row #' object. An mmkin row object is essentially a list of mkinfit objects that #' have been obtained by fitting the same model to a list of datasets. +#' +#' Note that the convergence of the nlme algorithms depends on the quality +#' of the data. In degradation kinetics, we often only have few datasets +#' (e.g. data for few soils) and complicated degradation models, which may +#' make it impossible to obtain convergence with nlme. #' #' @param model An [mmkin] row object. #' @param data Ignored, data are taken from the mmkin model @@ -88,11 +93,10 @@ get_deg_func <- function() { #' # f_nlme_sfo_sfo_ff <- nlme(f_2["SFO-SFO-ff", ]) #' #plot(f_nlme_sfo_sfo_ff) #' -#' # With the log-Cholesky parameterization, this converges in 11 -#' # iterations and around 100 seconds, but without tweaking control -#' # parameters (with pdDiag, increasing the tolerance and pnlsMaxIter was -#' # necessary) -#' f_nlme_dfop_sfo <- nlme(f_2["DFOP-SFO", ]) +#' # For the following, we need to increase pnlsMaxIter and the tolerance +#' # to get convergence +#' f_nlme_dfop_sfo <- nlme(f_2["DFOP-SFO", ], +#' control = list(pnlsMaxIter = 120, tolerance = 5e-4)) #' #' plot(f_nlme_dfop_sfo) #' @@ -112,22 +116,18 @@ get_deg_func <- function() { #' print(f_nlme_dfop_tc) #' } #' -#' f_2_obs <- mmkin(list("SFO-SFO" = m_sfo_sfo, -#' "DFOP-SFO" = m_dfop_sfo), -#' ds_2, quiet = TRUE, error_model = "obs") +#' f_2_obs <- update(f_2, error_model = "obs") #' f_nlme_sfo_sfo_obs <- nlme(f_2_obs["SFO-SFO", ]) #' print(f_nlme_sfo_sfo_obs) -#' f_nlme_dfop_sfo_obs <- nlme(f_2_obs["DFOP-SFO", ]) +#' f_nlme_dfop_sfo_obs <- nlme(f_2_obs["DFOP-SFO", ], +#' control = list(pnlsMaxIter = 120, tolerance = 5e-4)) #' -#' f_2_tc <- mmkin(list("SFO-SFO" = m_sfo_sfo, -#' "DFOP-SFO" = m_dfop_sfo), -#' ds_2, quiet = TRUE, error_model = "tc") -#' # f_nlme_sfo_sfo_tc <- nlme(f_2_tc["SFO-SFO", ]) # stops with error message -#' f_nlme_dfop_sfo_tc <- nlme(f_2_tc["DFOP-SFO", ]) -#' # We get warnings about false convergence in the LME step in several iterations -#' # but as the last such warning occurs in iteration 25 and we have 28 iterations -#' # we can ignore these -#' anova(f_nlme_dfop_sfo, f_nlme_dfop_sfo_obs, f_nlme_dfop_sfo_tc) +#' f_2_tc <- update(f_2, error_model = "tc") +#' # f_nlme_sfo_sfo_tc <- nlme(f_2_tc["SFO-SFO", ]) # No convergence with 50 iterations +#' # f_nlme_dfop_sfo_tc <- nlme(f_2_tc["DFOP-SFO", ], +#' # control = list(pnlsMaxIter = 120, tolerance = 5e-4)) # Error in X[, fmap[[nm]]] <- gradnm +#' +#' anova(f_nlme_dfop_sfo, f_nlme_dfop_sfo_obs) #' #' } nlme.mmkin <- function(model, data = "auto", |