Resources
Apps
Bend your (sp)line: A shiny app developed by TG2
A shiny app to visualize the possible different functional forms which can be modelled with fractional polynomials, natural (restricted cubic) splines, or linear B-splines. For eleven different explanatory variables, you can visualise non-linear effects applying these popular methods of non-linear modeling. You can visualise first- or second-degree fractional polynomials, linear B-splines with up to 4 degrees of freedom and natural splines with up to 3 degrees of freedom.
R packages and other resources provided by TG2 members
Regression Modeling Strategies
Regression modeling, testing, estimation, validation, graphics, prediction, and typesetting by storing enhanced model design attributes in the fit. ‘rms’ is a collection of functions that assist with and streamline modeling.
- R package: rms
- Online book companion: https://hbiostat.org/doc/rms/book/
Multivariable Fractional Polynomials (MFP)
MFP is a pragmatic procedure to create a multivariable model with the twin aims of selecting important variables and determining a suitable functional form for continuous predictors.
Resources:
- R-package mfp
- R-package mfp2
- MFP website with explanations, examples, software, links, etc.: https://mfp.imbi.uni-freiburg.de/software
mfp2 is a more recent package which includes several options and some extensions. A detailed vignette should help working with mfp2.
Augmented Backward Elimination
Augmented backward elimination combines significance or information based criteria with the change in estimate to either select the optimal model for prediction purposes or to serve as a tool to obtain a practically sound, highly interpretable model. More details can be found in Dunkler et al. (2014).
- R package: abe Performs augmented backward elimination and checks the stability of the obtained model.
Tetris-plot
The Tetris-plot is a tool to visualize the instability of models obtained by variable selection.
- Github website with R package (development version): https://gsk-biostatistics.github.io/tetrisplot