Methods for Evaluating Principal Surrogates of Treatment Response

Contains the core methods for the evaluation of principal surrogates in a single clinical trial. Provides a flexible interface for defining models for the risk given treatment and the surrogate, the models for integration over the missing counterfactual surrogate responses, and the estimation methods. Estimated maximum likelihood and pseudo-score can be used for estimation, and the bootstrap for inference. A variety of post-estimation summary methods are provided, including print, summary, plot, and testing.

MCMC Sampling of Bayesian Linear Models via Summary Statistics

Methods for generating Markov Chain Monte Carlo (MCMC) posterior samples of Bayesian linear regression model parameters that require only summary statistics of data as input. Summary statistics are useful for systems with very limited amounts of physical memory. The package provides two functions: one function that computes summary statistics of data and one function that carries out the MCMC posterior sampling for Bayesian linear regression models where summary statistics are used as input. The function read.regress.data.ff utilizes the R package 'ff' to handle data sets that are too large to fit into a user's physical memory, by reading in data in chunks. See Miroshnikov, Savel'ev and Conlon (2015) .

Curve Registration for Exponential Family Functional Data

A method for performing joint registration and functional principal component analysis for curves (functional data) that are generated from exponential family distributions. This mainly implements the algorithms described in 'Wrobel et al. (2019)' and further adapts them to potentially incomplete curves where (some) curves are not observed from the beginning and/or until the end of the common domain. Curve registration can be used to better understand patterns in functional data by separating curves into phase and amplitude variability. This software handles both binary and continuous functional data, and is especially applicable in accelerometry and wearable technology.

Diagnostics for Confounding of Time-Varying and Other Joint Exposures

Implements three covariate-balance diagnostics for time-varying confounding and selection-bias in complex longitudinal data, as described in Jackson (2016) and Jackson (2019) . Diagnostic 1 assesses measured confounding/selection-bias, diagnostic 2 assesses exposure-covariate feedback, and diagnostic 3 assesses residual confounding/selection-bias after inverse probability weighting or propensity score stratification. All diagnostics appropriately account for exposure history, can be adapted to assess a particular depth of covariate history, and can be implemented in right-censored data. Balance assessments can be obtained for all times, selected-times, or averaged across person-time. The balance measures are reported as tables or plots. These diagnostics can be applied to the study of multivariate exposures including time-varying exposures, direct effects, interaction, and censoring.

Causal Inference Modeling for Estimation of Causal Effects

Provides an array of statistical models common in causal inference such as standardization, IP weighting, propensity matching, outcome regression, and doubly-robust estimators. Estimates of the average treatment effects from each model are given with the standard error and a 95% Wald confidence interval (Hernan, Robins (2020) < https://miguelhernan.org/whatifbook/>).

Access Meetup Data

Provides programmatic access to the 'Meetup' 'GraphQL' API (< https://www.meetup.com/graphql/>), enabling users to retrieve information about groups, events, and members from 'Meetup' (< https://www.meetup.com/>). Supports authentication via 'OAuth2' and includes functions for common queries and data manipulation tasks.

A Package for Processing Lexical Response Data

Lexical response data is a package that can be used for processing cued-recall, free-recall, and sentence responses from memory experiments.

Kernel Balancing

Provides a weighting approach that employs kernels to make one group have a similar distribution to another group on covariates. This method matches not only means or marginal distributions but also higher-order transformations implied by the choice of kernel. 'kbal' is applicable to both treatment effect estimation and survey reweighting problems. Based on Hazlett, C. (2020) "Kernel Balancing: A flexible non-parametric weighting procedure for estimating causal effects." Statistica Sinica. < https://www.researchgate.net/publication/299013953_Kernel_Balancing_A_flexible_non-parametric_weighting_procedure_for_estimating_causal_effects>.

Inference for Functions of Multinomial Parameters

We consider the problem where we observe k vectors (possibly of different lengths), each representing an independent multinomial random vector. For a given function that takes in the concatenated vector of multinomial probabilities and outputs a real number, this is a Monte Carlo estimation procedure of an exact p-value and confidence interval. The resulting inference is valid even in small samples, when the parameter is on the boundary, and when the function is not differentiable at the parameter value, all situations where asymptotic methods and the bootstrap would fail. For more details see Sachs, Fay, and Gabriel (2025) .

Regression Standardization for Causal Inference

Contains more modern tools for causal inference using regression standardization. Four general classes of models are implemented; generalized linear models, conditional generalized estimating equation models, Cox proportional hazards models, and shared frailty gamma-Weibull models. Methodological details are described in Sjölander, A. (2016) . Also includes functionality for doubly robust estimation for generalized linear models in some special cases, and the ability to implement custom models.