Ray-Based Mapping and Visualization of Level Sets (Excursion Sets)

An (upper) level set of a function is the set of inputs for which the function value is at or above a specified threshold. (Also called an excursion set). Applications of level sets include confidence or credible regions for parameters of statistical models, where the function is the likelihood or posterior density; regions where classification rules assign high probability to a given class; and scientific or engineering models where one is interested in input regions for which model output is above a threshold. This package maps out the boundary of a level set by finding its intersections with collections of 1-dimensional rays, generalizing a proposal by Kim and Lindsay (Statistica Sinica 21:923-948, 2011). Tools are provided to generate rays, find intersections, and visualize results. The package makes few assumptions about the studied function: it may be discontinuous, it may have a complicated feasible region, and the target level set may be non-convex or have multiple, disconnected parts. Vignettes describe package usage and show examples with two to five input space dimensions.

Stratified Randomized Experiments

Estimate average treatment effects (ATEs) in stratified randomized experiments. 'sreg' supports a wide range of stratification designs, including matched pairs, n-tuple designs, and larger strata with many units — possibly of unequal size across strata. 'sreg' is designed to accommodate scenarios with multiple treatments and cluster-level treatment assignments, and accommodates optimal linear covariate adjustment based on baseline observable characteristics. 'sreg' computes estimators and standard errors based on Bugni, Canay, Shaikh (2018) ; Bugni, Canay, Shaikh, Tabord-Meehan (2024+) ; Jiang, Linton, Tang, Zhang (2023+) ; Bai, Jiang, Romano, Shaikh, and Zhang (2024) ; Bai (2022) ; Bai, Romano, and Shaikh (2022) ; Liu (2024+) ; and Cytrynbaum (2024) .

Easily Install and Load the 'Tidyverse'

The 'tidyverse' is a set of packages that work in harmony because they share common data representations and 'API' design. This package is designed to make it easy to install and load multiple 'tidyverse' packages in a single step. Learn more about the 'tidyverse' at < https://www.tidyverse.org>.

Assessing and Visualizing Fragility of Clinical Results

A collection of user-friendly functions for assessing fragility of clinical results with binary and survival outcomes. For binary outcomes, the package assesses and visualizes fragility of individual studies (Walsh et al., 2014 ; Lin, 2021 ), conventional pairwise meta-analyses (Atal et al., 2019 ), and network meta-analyses of multiple treatments with binary outcomes (Xing et al., 2020 ). The functions for binary outcomes are designed to: 1) calculate the fragility index (i.e., the minimal event status modifications that can alter the significance or non-significance of the original result) and fragility quotient (i.e., fragility index divided by sample size) at a specific significance level; 2) give the cases of event status modifications for altering the result's significance or non-significance and visualize these cases; 3) visualize the trend of statistical significance as event status is modified; 4) efficiently derive fragility indexes and fragility quotients at multiple significance levels, and visualize the relationship between these fragility measures against the significance levels; and 5) calculate fragility indexes and fragility quotients of multiple datasets (e.g., a collection of clinical trials or meta-analyses) and produce plots of their overall distributions. For survival outcomes, the package implements the event status modification method based on the log-rank test described by Xing et al. (2026 ). It calculates the fragility index and fragility quotient for two-group studies with right-censored data, modifying event status in one or both groups while preserving follow-up times and group assignments. Results include the sequence of modifications, corresponding p-values, and an S3 print method. The outputs from these functions may inform the robustness of clinical results in terms of statistical significance and aid the interpretation of fragility measures. The usage of this package is illustrated in Lin et al. (2023 ) and detailed in Lin and Chu (2022 ).

Biological Condition Gradient, Calculator

Functions to calculate Biological Condition Gradient (BCG) using input files with one row per sample with metric values and site classes as columns. A second file with the BCG Rules (example included) to define the memberships is also needed. The three main functions convert metric scores to metric memberships following fuzzy set BCG Rules (BCG.Level.Assignment), combine metric memberships to level memberships according to BCG Rules (BCG.Level.Membership), and then assign a BCG primary and secondary level based on level memberships (BCG.Level.Assignment). Originally developed as a package for use with BCG for Puget Lowland/Willamette Valley but has been further enhanced for use with multiple communities (benthic macroinvertebrates, fish, periphyton, or coral) and different rule sets. Oregon and Washington reference - "Stamp, J. and J. Gerritsen. 2018. Calibration of the Biological Condition Gradient (BCG) for Macroinvertebrate Assemblages in Puget Lowland/Willamette Valley Freshwater Wadeable Streams. Prepared by Tetra Tech for the US EPA Office of Water, Office of Science and Technology and US EPA Region 10." BCG process documentation - "USEPA. 2016. A Practitioner’s Guide to the Biological Condition Gradient - A Framework to Describe Incremental Change in Aquatic Ecosystems. EPA 842-R-16-001. Office of Science and Technology, Washington, DC 20460."

Estimating Length-Based Indicators for Fish Stock

Provides tools for estimating length-based indicators from length frequency data to assess fish stock status and manage fisheries sustainably. Implements methods from Cope and Punt (2009) for data-limited stock assessment and Froese (2004) for detecting overfishing using simple indicators. Key functions include: FrequencyTable(): Calculate the frequency table from the collected and also the extract the length frequency data from the frequency table with the upper length_range. A numeric value specifying the bin width for class intervals. If not provided, the bin width is automatically calculated using Wang (2020) formula. FreqTM(): Creates a frequency distribution table for fish length data across multiple months using a consistent length class structure. The bin width is determined by either a custom value or Wang's formula, applied uniformly across all months. The function dynamically detects and renames columns to 'Month' and 'Length' from the input dataframe. The maximum observed length is included as part of the last class, with the upper bound set to the smallest multiple of the bin width greater than or equal to the maximum length. Months can be converted to dates using a configurable day and year, with dates assigned sequentially in 'day.month.year' format (e.g., 15.01.26). FishPar(): Calculates length-based indicators (LBIs) proposed by Froese (2004) such as the percentage of mature fish (Pmat), percentage of optimal length fish (Popt), percentage of mega spawners (Pmega), and the sum of these as Pobj. This function also estimates confidence intervals for different lengths, visualizes length frequency distributions, and provides data frames containing calculated values. FishSS(): Makes decisions based on input from Cope and Punt (2009) and parameters calculated by FishPar() (e.g., Pobj, Pmat, Popt, LM_ratio) to determine stock status as target spawning biomass (TSB40) and limit spawning biomass (LSB25), and selectivity. LWR(): Fits and visualizes length-weight relationships using linear regression, with options for log-transformation and customizable plotting.

Extended Model Formulas

Infrastructure for extended formulas with multiple parts on the right-hand side and/or multiple responses on the left-hand side (see ).

Automatic Toolkit for Construction, Optimization, Scoring and Simulation of Forced-Choice Tests

Forced-choice (FC) response has gained increasing popularity and interest for its resistance to faking when well-designed (Cao & Drasgow, 2019 ). To established well-designed FC scales, typically each item within a block should measure different trait and have similar level of social desirability (Zhang et al., 2020 ). Recent study also suggests the importance of high inter-item agreement of social desirability between items within a block (Pavlov et al., 2021 ). In addition to this, FC developers may also need to maximize factor loading differences (Brown & Maydeu-Olivares, 2011 ) or minimize item location differences (Cao & Drasgow, 2019 ) depending on scoring models. Decision of which items should be assigned to the same block, also called as item pairing, is thus critical to the quality of an FC test. Because such pairing process often requires researchers to meet multiple objectives, manual pairing becomes impractical or even not feasible once the number of latent traits and/or number of items per elevates. To address these problems, autoFC is developed as a automatic and efficient tool for facilitating the automatic construction of FC tests (Li et al., 2022 ), essentially exempting users from the burden of manual item pairing. Given characteristics of each item (and item responses), FC measures can be constructed either automatically based on user-defined pairing criteria and weights, or based on exact specifications of each block (i.e., blueprint; see Li et al., 2025 ). Users can also generate simulated responses based on the Thurstonian Item Response Theory model (Brown & Maydeu-Olivares, 2011 ) and predict trait scores of simulated/actual respondents based on an estimated model.

Manage Environment Specific Configuration Values

Manage configuration values across multiple environments (e.g. development, test, production). Read values using a function that determines the current environment and returns the appropriate value.

Tidying Methods for Mixed Models

Convert fitted objects from various R mixed-model packages into tidy data frames along the lines of the 'broom' package. The package provides three S3 generics for each model: tidy(), which summarizes a model's statistical findings such as coefficients of a regression; augment(), which adds columns to the original data such as predictions, residuals and cluster assignments; and glance(), which provides a one-row summary of model-level statistics.