The HJ-Biplot is a multivariate method that represents high-dimensional data in a low-dimensional subspace, capturing most of the informationâ s variability in just a few dimensions. This package implements three new regularized versions of the HJ-Biplot: Ridge, LASSO, and Elastic Net. These versions introduce restrictions that shrink or zero-out variable weights to improve interpretability based on regularization theory. All methods provide graphical representations using ggplot2'.

Training and validation of a custom (or data-driven) Structural Equation Models using Deep Neural Networks or Machine Learning algorithms, which extend the fitting procedures of the SEMgraph R package <doi:10.32614/CRAN.package.SEMgraph>.

This R package introduces Weighted Mean SHapley Additive exPlanations (WMSHAP), an innovative method for calculating SHAP values for a grid of fine-tuned base-learner machine learning models as well as stacked ensembles, a method not previously available due to the common reliance on single best-performing models. By integrating the weighted mean SHAP values from individual base-learners comprising the ensemble or individual base-learners in a tuning grid search, the package weights SHAP contributions according to each model's performance, assessed by multiple either R squared (for both regression and classification models). alternatively, this software also offers weighting SHAP values based on the area under the precision-recall curve (AUCPR), the area under the curve (AUC), and F2 measures for binary classifiers. It further extends this framework to implement weighted confidence intervals for weighted mean SHAP values, offering a more comprehensive and robust feature importance evaluation over a grid of machine learning models, instead of solely computing SHAP values for the best model. This methodology is particularly beneficial for addressing the severe class imbalance (class rarity) problem by providing a transparent, generalized measure of feature importance that mitigates the risk of reporting SHAP values for an overfitted or biased model and maintains robustness under severe class imbalance, where there is no universal criteria of identifying the absolute best model. Furthermore, the package implements hypothesis testing to ascertain the statistical significance of SHAP values for individual features, as well as comparative significance testing of SHAP contributions between features. Additionally, it tackles a critical gap in feature selection literature by presenting criteria for the automatic feature selection of the most important features across a grid of models or stacked ensembles, eliminating the need for arbitrary determination of the number of top features to be extracted. This utility is invaluable for researchers analyzing feature significance, particularly within severely imbalanced outcomes where conventional methods fall short. Moreover, it is also expected to report democratic feature importance across a grid of models, resulting in a more comprehensive and generalizable feature selection. The package further implements a novel method for visualizing SHAP values both at subject level and feature level as well as a plot for feature selection based on the weighted mean SHAP ratios.

This package performs multivariate nonparametric regression/classification by the method of sieves (using orthogonal basis). The method is suitable for moderate high-dimensional features (dimension < 100). The l1-penalized sieve estimator, a nonparametric generalization of Lasso, is adaptive to the feature dimension with provable theoretical guarantees. We also include a nonparametric stochastic gradient descent estimator, Sieve-SGD, for online or large scale batch problems. Details of the methods can be found in: <arXiv:2206.02994> <arXiv:2104.00846><arXiv:2310.12140>.

Median-of-means is a generic yet powerful framework for scalable and robust estimation. A framework for Bayesian analysis is called M-posterior, which estimates a median of subset posterior measures. For general exposition to the topic, see the paper by Minsker (2015) <doi:10.3150/14-BEJ645>.

This package provides functions to perform simulations of ANOVA designs of up to three factors. Calculates the observed power and average observed effect size for all main effects and interactions in the ANOVA, and all simple comparisons between conditions. Includes functions for analytic power calculations and additional helper functions that compute effect sizes for ANOVA designs, observed error rates in the simulations, and functions to plot power curves. Please see Lakens, D., & Caldwell, A. R. (2021). "Simulation-Based Power Analysis for Factorial Analysis of Variance Designs". <doi:10.1177/2515245920951503>.

Computes the sBIC for various singular model collections including: binomial mixtures, factor analysis models, Gaussian mixtures, latent forests, latent class analyses, and reduced rank regressions.

Method to estimate the spatial influence scales of landscape variables on a response variable. The method is based on Chandler and Hepinstall-Cymerman (2016) Estimating the spatial scales of landscape effects on abundance, Landscape ecology, 31: 1383-1394, <doi:10.1007/s10980-016-0380-z>.

Compute the position of the sun, and local solar time using Meeus formulae. Compute day and/or night length using different twilight definitions or arbitrary sun elevation angles. This package is part of the r4photobiology suite, Aphalo, P. J. (2015) <doi:10.19232/uv4pb.2015.1.14>. Algorithms from Meeus (1998, ISBN:0943396611).

Uses logistic regression to model the probability of detection as a function of covariates. This model is then used with observational survey data to estimate population size, while accounting for uncertain detection. See Steinhorst and Samuel (1989).

Computes the sit coefficient between two vectors x and y, possibly all paired coefficients for a matrix. The reference for the methods implemented here is Zhang, Yilin, Canyi Chen, and Liping Zhu. 2022. "Sliced Independence Test." Statistica Sinica. <doi:10.5705/ss.202021.0203>. This package incorporates the Galton peas example.

Perform association test within linear mixed model framework using score test integrated with Empirical Bayes for genome-wide association study. Firstly, score test was conducted for each marker under linear mixed model framework, taking into account the genetic relatedness and population structure. And then all the potentially associated markers were selected with a less stringent criterion. Finally, all the selected markers were placed into a multi-locus model to identify the true quantitative trait nucleotide.

Easy-to-use interface to X-13-ARIMA-SEATS, the seasonal adjustment software by the US Census Bureau. It offers full access to almost all options and outputs of X-13, including X-11 and SEATS, automatic ARIMA model search, outlier detection and support for user defined holiday variables, such as Chinese New Year or Indian Diwali. A graphical user interface can be used through the seasonalview package. Uses the X-13-binaries from the x13binary package.

An efficient implementation of Scalable Bayesian Rule Lists Algorithm, a competitor algorithm for decision tree algorithms; see Hongyu Yang, Cynthia Rudin, Margo Seltzer (2017) <https://proceedings.mlr.press/v70/yang17h.html>. It builds from pre-mined association rules and have a logical structure identical to a decision list or one-sided decision tree. Fully optimized over rule lists, this algorithm strikes practical balance between accuracy, interpretability, and computational speed.

Implementation of SPECS, your favourite Single-Equation Penalized Error-Correction Selector developed in Smeekes and Wijler (2021) <doi:10.1016/j.jeconom.2020.07.021>. SPECS provides a fully automated estimation procedure for large and potentially (co)integrated datasets. The dataset in levels is converted to a conditional error-correction model, either by the user or by means of the functions included in this package, and various specialised forms of penalized regression can be applied to the model. Automated options for initializing and selecting a sequence of penalties, as well as the construction of penalty weights via an initial estimator, are available. Moreover, the user may choose from a number of pre-specified deterministic configurations to further simplify the model building process.

Efficient procedure for fitting regularization paths between L1 and L0, using the MC+ penalty of Zhang, C.H. (2010)<doi:10.1214/09-AOS729>. Implements the methodology described in Mazumder, Friedman and Hastie (2011) <DOI: 10.1198/jasa.2011.tm09738>. Sparsenet computes the regularization surface over both the family parameter and the tuning parameter by coordinate descent.

This package provides tools for setting up ("design"), conducting, and evaluating large-scale simulation studies with graphics and tables, including parallel computations.

This package performs exploratory factor analysis on language model embeddings of psychological scale items. Embeds item text with sentence transformers or other language models, transforms the embeddings into item-by-item similarity matrices, and extracts latent factor structure via standard exploratory factor analysis. Supports embedding-adapted parallel analysis, several similarity transforms (atomic reversed, SQuID centering, mean-centered Pearson), and fit diagnostics tailored to embedding matrices (TEFI, RMSR, CAF, McDonald's omega). The underlying methods are documented with full citations in the corresponding function help pages. Returns objects compatible with psych and EFAtools workflows.

This package provides functions to generate K-fold cross validation (CV) folds and CV test error estimates that take into account how a survey dataset's sampling design was constructed (SRS, clustering, stratification, and/or unequal sampling weights). You can input linear and logistic regression models, along with data and a type of survey design in order to get an output that can help you determine which model best fits the data using K-fold cross validation. Our paper on "K-Fold Cross-Validation for Complex Sample Surveys" by Wieczorek, Guerin, and McMahon (2022) <doi:10.1002/sta4.454> explains why differing how we take folds based on survey design is useful.

This package provides methods that use flexible variants of multidimensional scaling (MDS) which incorporate parametric nonlinear distance transformations and trade-off the goodness-of-fit fit with structure considerations to find optimal hyperparameters, also known as structure optimized proximity scaling (STOPS) (Rusch, Mair & Hornik, 2023,<doi:10.1007/s11222-022-10197-w>). The package contains various functions, wrappers, methods and classes for fitting, plotting and displaying different 1-way MDS models with ratio, interval, ordinal optimal scaling in a STOPS framework. These cover essentially the functionality of the package smacofx, including Torgerson (classical) scaling with power transformations of dissimilarities, SMACOF MDS with powers of dissimilarities, Sammon mapping with powers of dissimilarities, elastic scaling with powers of dissimilarities, spherical SMACOF with powers of dissimilarities, (ALSCAL) s-stress MDS with powers of dissimilarities, r-stress MDS, MDS with powers of dissimilarities and configuration distances, elastic scaling powers of dissimilarities and configuration distances, Sammon mapping powers of dissimilarities and configuration distances, power stress MDS (POST-MDS), approximate power stress, Box-Cox MDS, local MDS, Isomap, curvilinear component analysis (CLCA), curvilinear distance analysis (CLDA) and sparsified (power) multidimensional scaling and (power) multidimensional distance analysis (experimental models from smacofx influenced by CLCA). All of these models can also be fit by optimizing over hyperparameters based on goodness-of-fit fit only (i.e., no structure considerations). The package further contains functions for optimization, specifically the adaptive Luus-Jaakola algorithm and a wrapper for Bayesian optimization with treed Gaussian process with jumps to linear models, and functions for various c-structuredness indices. Hyperparameter optimization can be done with a number of techniques but we recommend either Bayesian optimization or particle swarm. For using "Kriging", users need to install a version of the archived DiceOptim R package.

The strip function deletes components of R model outputs that are useless for specific purposes, such as predict[ing], print[ing], summary[izing], etc.

This package implements a sequential imputation framework using Bayesian Mixed-Effects Trees ('SBMTrees') for handling missing data in longitudinal studies. The package supports a variety of models, including non-linear relationships and non-normal random effects and residuals, leveraging Dirichlet Process priors for increased flexibility. Key features include handling Missing at Random (MAR) longitudinal data, imputation of both covariates and outcomes, and generating posterior predictive samples for further analysis. The methodology is designed for applications in epidemiology, biostatistics, and other fields requiring robust handling of missing data in longitudinal settings.

Computes the probability of a set of species abundances of a single or multiple samples of individuals with one or more guilds under a mainland-island model. One must specify the mainland (metacommunity) model and the island (local) community model. It assumes that species fluctuate independently. The package also contains functions to simulate under this model. See Haegeman, B. & R.S. Etienne (2017). A general sampling formula for community structure data. Methods in Ecology & Evolution 8: 1506-1519 <doi:10.1111/2041-210X.12807>.

Generates artificial point patterns marked by their spatial and temporal signatures. The resulting point cloud may exhibit inherent interactions between both signatures. The simulation integrates microsimulation (Holm, E., (2017)<doi:10.1002/9781118786352.wbieg0320>) and agent-based models (Bonabeau, E., (2002)<doi:10.1073/pnas.082080899>), beginning with the configuration of movement characteristics for the specified agents (referred to as walkers') and their interactions within the simulation environment. These interactions (Quaglietta, L. and Porto, M., (2019)<doi:10.1186/s40462-019-0154-8>) result in specific spatiotemporal patterns that can be visualized, analyzed, and used for various analytical purposes. Given the growing scarcity of detailed spatiotemporal data across many domains, this package provides an alternative data source for applications in social and life sciences.