Current versions

saemix 3.0

News

In the works!

Work is currently undergoing to improve saemix in the following directions:

• improving the handling of covariates: centering and transforming continuous covariates, automatically creating dummy covariates to allow for more than two categories for categorical covariates
• multi-response models
• time-to-event models with competing risks
• joint models with a longitudinal response and a survival response
• flexible error models for continuous responses
• hidden Markov models
• estimation of parameter uncertainty by bootstrap and SIR methods
• left censored data
• mixture models

A development version is available on github and you are more than welcome to join the team (please drop a line to emmanuelle.comets@inserm.fr).

Changelog

A full changelog is available in the installed package (please refer to the file called CHANGES).

Version 3.0, February 2022

1. Extension to non-continuous data models
   • New argument in saemixModel to differentiate between continuous (modelType = "structural") and noncontinuous (modelType = "likelihood")
   • New examples have been included in the package to showcase discrete response models (see the userguide for more information on lung.saemix, toenail.saemix, knee.saemix, rapi.saemix)
2. Main changes
   • Plot functions for models have been added to help choose adequate initial values for model parameters
   • Implementation of a covariate model building algorithm based on the BICc proposed for NLMEM
   • Verbosity has been considerably reduced by setting displayProgress to FALSE and removing most of the messages by default (they can be activated through arguments to the call).
   • Improved diagnostic graphs using the npde package: as a result, saemix now depends on npde (3.2), ggplot2, grid and gridExtra
3. Bugfixes: see CHANGELOG for details
   • Corrected a mistake in the inverse transformation for the logit function
   • Corrected a bug where plot.type="population.fit" gave the same result as plot.type="individual.fit"

Version 2.4, February 2021

1. Main changes
   • change in the form of the combined error model
   • computation of the correlation and their SE along with the variance parameters
2. Bugfixes: correction in the computation of the individual conditional distributions

Version 2.3, October 2019

1. Update following a change in the compiler on CRAN

Version 2.2, October 2018

1. Minor changes to SaemixData, SaemixModel and SaemixObject objects
2. Bugfixes
   • computation of the Fisher Information Matrix recoded to fix an underestimation in the standard errors of the residual error terms
   • the FIM is now given in the results (slot fim of the results element) instead of -FIM

Version 2.1, August 2017

1. Version of the saemix package to CRAN that refers back to the JSS manuscript: Comets E, Lavenu A, Lavielle M. Parameter Estimation in Nonlinear Mixed Effect Models Using saemix, an R Implementation of the SAEM Algorithm. Journal of Statistical Software (2017), 80(3):1-41. Comets et al. JSS 2017

Version 2.0, September 2015

1. New features
   • the predict() function has been changed to return model predictions (see help file for details)
   • new extractor functions have been programmed: logLik, residuals, AIC, BIC

Version 1.2, February 2014

1. Bugfixes
   • Corrected a bug which caused the run to fail when the model had only one random effect
   • Explicit error message when attempting to create a model without random effects
   • Minor changes

Version 1.1, February 2013

1. New features
   • Binary covariates can now be entered directly as factors; if entered as strings or numbers, they will be converted to factors. Using print() on the result of saemix will show which is the reference class.
   • A function subset has been defined for a SaemixData object.
   • A function logLik has been defined to extract the log-likelihood from a SaemixObject object.
2. Bugfixes: see details in CHANGES
   • Corrected a bug where plot.type="population.fit" gave the same result as plot.type="individual.fit"
3. Changes
   • Covariates can now be entered as categorical (eg using Male/Female levels for gender instead of 0/1)
   • An error capture was implemented to avoid the fit failing because the convergence plots cannot be plotted, which may happen when the number of parameters to estimate is very large or the size of the plotting region too small (this can easily occur when using Rstudio with a small plot window). The run would fail with the following error message (in English):
     Error in plot.new(): figure margins too large
     Now an error message will be printed out, and the graphs will not be created during the fit. However, they can still be obtained after the fit using the plot.type="convergence" argument and a suitable mfrow argument (eg mfrow=c(1,1) will plot the convergence graph for each parameter on a new page).
   • Tests have been added when matching the covariate model specified with saemixModel and the data provided through saemixData
   • when there are less covariates than those specified in the model, only the first lines of the covariate model matrix are used (as many lines as there are covariates in the model
   • when there are more covariates than those specified in the model, an absence of relationship is assumed for the additional covariates.
   • Missing covariates previously caused the fit to fail with a unenlightening error message. This has been corrected: now when covariates enter the model, lines with missing covariates are removed; this may cause some individuals to be removed from the dataset altogether if the corresponding covariates are missing entirely.

Version 0.96, July 2011

First version uploaded to CRAN