Causal Mediation Analysis

We implement parametric and non parametric mediation analysis. This package performs the methods and suggestions in Imai, Keele and Yamamoto (2010) , Imai, Keele and Tingley (2010) , Imai, Tingley and Yamamoto (2013) , and Imai and Yamamoto (2013) . In addition to the estimation of causal mediation effects, the software also allows researchers to conduct sensitivity analysis for certain parametric models.

Bayesian Applied Regression Modeling via Stan

Estimates previously compiled regression models using the 'rstan' package, which provides the R interface to the Stan C++ library for Bayesian estimation. Users specify models via the customary R syntax with a formula and data.frame plus some additional arguments for priors.

Visualization and Imputation of Missing Values

Provides methods for imputation and visualization of missing values. It includes graphical tools to explore the amount, structure and patterns of missing and/or imputed values, supporting exploratory data analysis and helping to investigate potential missingness mechanisms (details in Alfons, Templ and Filzmoser, . The quality of imputations can be assessed visually using a wide range of univariate, bivariate and multivariate plots. The package further provides several imputation methods, including efficient implementations of k-nearest neighbour and hot-deck imputation (Kowarik and Templ 2013, , iterative robust model-based multiple imputation (Templ 2011, ; Templ 2023, ), and machine learning–based approaches such as robust GAM-based multiple imputation (Templ 2024, ) as well as gradient boosting (XGBoost) and transformer-based methods (Niederhametner et al., ). General background and practical guidance on imputation are provided in the Springer book by Templ (2023) .

A Common API to Modeling and Analysis Functions

A common interface is provided to allow users to specify a model without having to remember the different argument names across different functions or computational engines (e.g. 'R', 'Spark', 'Stan', 'H2O', etc).

Tidy Tuning Tools

The ability to tune models is important. 'tune' contains functions and classes to be used in conjunction with other 'tidymodels' packages for finding reasonable values of hyper-parameters in models, preprocessing methods, and post-processing steps.

Estimated Marginal Means, aka Least-Squares Means

Obtain estimated marginal means (EMMs) for many linear, generalized linear, and mixed models. Compute contrasts or linear functions of EMMs, trends, and comparisons of slopes. Plots and other displays. Least-squares means are discussed, and the term "estimated marginal means" is suggested, in Searle, Speed, and Milliken (1980) Population marginal means in the linear model: An alternative to least squares means, The American Statistician 34(4), 216-221 .

Bayesian Regression Models using 'Stan'

Fit Bayesian generalized (non-)linear multivariate multilevel models using 'Stan' for full Bayesian inference. A wide range of distributions and link functions are supported, allowing users to fit -- among others -- linear, robust linear, count data, survival, response times, ordinal, zero-inflated, hurdle, and even self-defined mixture models all in a multilevel context. Further modeling options include both theory-driven and data-driven non-linear terms, auto-correlation structures, censoring and truncation, meta-analytic standard errors, and quite a few more. In addition, all parameters of the response distribution can be predicted in order to perform distributional regression. Prior specifications are flexible and explicitly encourage users to apply prior distributions that actually reflect their prior knowledge. Models can easily be evaluated and compared using several methods assessing posterior or prior predictions. References: Bürkner (2017) ; Bürkner (2018) ; Bürkner (2021) ; Carpenter et al. (2017) .

Distributed Lag Non-Linear Models

Collection of functions for distributed lag linear and non-linear models.

Functional Data Analysis and Empirical Dynamics

A versatile package that provides implementation of various methods of Functional Data Analysis (FDA) and Empirical Dynamics. The core of this package is Functional Principal Component Analysis (FPCA), a key technique for functional data analysis, for sparsely or densely sampled random trajectories and time courses, via the Principal Analysis by Conditional Estimation (PACE) algorithm. This core algorithm yields covariance and mean functions, eigenfunctions and principal component (scores), for both functional data and derivatives, for both dense (functional) and sparse (longitudinal) sampling designs. For sparse designs, it provides fitted continuous trajectories with confidence bands, even for subjects with very few longitudinal observations. PACE is a viable and flexible alternative to random effects modeling of longitudinal data. There is also a Matlab version (PACE) that contains some methods not available on fdapace and vice versa. Updates to fdapace were supported by grants from NIH Echo and NSF DMS-1712864 and DMS-2014626. Please cite our package if you use it (You may run the command citation("fdapace") to get the citation format and bibtex entry). References: Wang, J.L., Chiou, J., Müller, H.G. (2016) ; Chen, K., Zhang, X., Petersen, A., Müller, H.G. (2017) .

Ordination and Multivariate Analysis for Ecology

A variety of ordination and community analyses useful in analysis of data sets in community ecology. Includes many of the common ordination methods, with graphical routines to facilitate their interpretation, as well as several novel analyses.