Fit Bayesian Gaussian graphical models. The methods are separated into two Bayesian approaches for inference: hypothesis testing and estimation. There are extensions for confirmatory hypothesis testing, comparing Gaussian graphical models, and node wise predictability. These methods were recently introduced in the Gaussian graphical model literature, including Williams (2019) <doi:10.31234/osf.io/x8dpr>, Williams and Mulder (2019) <doi:10.31234/osf.io/ypxd8>, Williams, Rast, Pericchi, and Mulder (2019) <doi:10.31234/osf.io/yt386>.

Explore neural networks in a layer oriented way, the framework is intended to give the user total control of the internals of a net without much effort. Use classes like PerceptronLayer to create a layer of Percetron neurons, and specify how many you want. The package does all the tricky stuff internally leaving you focused in what you want. I wrote this package during a neural networks course to help me with the problem set.

Allows to estimate and test high-dimensional mediation effects based on advanced mediator screening and penalized regression techniques. Methods used in the package refer to Zhang H, Zheng Y, Zhang Z, Gao T, Joyce B, Yoon G, Zhang W, Schwartz J, Just A, Colicino E, Vokonas P, Zhao L, Lv J, Baccarelli A, Hou L, Liu L. Estimating and Testing High-dimensional Mediation Effects in Epigenetic Studies. Bioinformatics. (2016) <doi:10.1093/bioinformatics/btw351>. PMID: 27357171.

The Gaussian location-scale regression model is a multi-predictor model with explanatory variables for the mean (= location) and the standard deviation (= scale) of a response variable. This package implements maximum likelihood and Markov chain Monte Carlo (MCMC) inference (using algorithms from Girolami and Calderhead (2011) <doi:10.1111/j.1467-9868.2010.00765.x> and Nesterov (2009) <doi:10.1007/s10107-007-0149-x>), a parametric bootstrap algorithm, and diagnostic plots for the model class.

Multivariate analysis, having functions that perform simple correspondence analysis (CA) and multiple correspondence analysis (MCA), principal components analysis (PCA), canonical correlation analysis (CCA), factorial analysis (FA), multidimensional scaling (MDS), linear (LDA) and quadratic discriminant analysis (QDA), hierarchical and non-hierarchical cluster analysis, simple and multiple linear regression, multiple factor analysis (MFA) for quantitative, qualitative, frequency (MFACT) and mixed data, biplot, scatter plot, projection pursuit (PP), grant tour method and other useful functions for the multivariate analysis.

This package provides methods to generate a design in the input space that sequentially fills the output space of a black-box function. The output space-filling designs are helpful in inverse design or feature-based modeling problems. See Wang, Shangkun, Adam P. Generale, Surya R. Kalidindi, and V. Roshan Joseph. (2024), Sequential designs for filling output spaces, Technometrics, 66, 65â 76. for details. This work is supported by U.S. National Foundation grant CMMI-1921646.

This package implements the Quantile Composite-based Path Modeling approach (Davino and Vinzi, 2016 <doi:10.1007/s11634-015-0231-9>; Dolce et al., 2021 <doi:10.1007/s11634-021-00469-0>). The method complements the traditional PLS Path Modeling approach, analyzing the entire distribution of outcome variables and, therefore, overcoming the classical exploration of only average effects. It exploits quantile regression to investigate changes in the relationships among constructs and between constructs and observed variables.

This package implements the methodological developments found in Hermes, van Heerwaarden, and Behrouzi (2023) <doi:10.48550/arXiv.2308.04325>, and allows for the statistical modeling of asymmetric between-location effects, as well as within-location effects using spatial autoregressive graphical models. The package allows for the generation of spatial weight matrices to capture asymmetric effects for strip-type intercropping designs, although it can handle any type of spatial data commonly found in other sciences.

Analyzes longitudinal data of HIV decline in patients on antiretroviral therapy using the canonical biphasic exponential decay model (pioneered, for example, by work in Perelson et al. (1997) <doi:10.1038/387188a0>; and Wu and Ding (1999) <doi:10.1111/j.0006-341X.1999.00410.x>). Model fitting and parameter estimation are performed, with additional options to calculate the time to viral suppression. Plotting and summary tools are also provided for fast assessment of model results.

Estimating the disparity between two groups based on the extended model of the Peters-Belson (PB) method. Our model is the first work on the longitudinal data, and also can set a varying variable to find the complicated association between other variables and the varying variable. Our work is an extension of the Peters-Belson method which was originally published in Peters (1941)<doi:10.1080/00220671.1941.10881036> and Belson (1956)<doi:10.2307/2985420>.

This package provides a convenient data set, a set of helper functions, and a benchmark function for economically (profit) driven wind farm layout optimization. This enables researchers in the field of the NP-hard (non-deterministic polynomial-time hard) problem of wind farm layout optimization to focus on their optimization methodology contribution and also provides a realistic benchmark setting for comparability among contributions. See Croonenbroeck, Carsten & Hennecke, David (2020) <doi:10.1016/j.energy.2020.119244>.

Calculate the win statistics (win ratio, net benefit and win odds) for prioritized multiple endpoints, plot the win statistics and win proportions over study time if at least one time-to-event endpoint is analyzed, and simulate datasets with dependent endpoints. The package can handle any type of outcomes (continuous, ordinal, binary, time-to-event) and allow users to perform stratified analysis, inverse probability of censoring weighting (IPCW) and inverse probability of treatment weighting (IPTW) analysis.

This package provides functions for the analysis of income distributions for subgroups of the population as defined by a set of variables like age, gender, region, etc. This entails a Kolmogorov-Smirnov test for a mixture distribution as well as functions for moments, inequality measures, entropy measures and polarisation measures of income distributions. This package thus aides the analysis of income inequality by offering tools for the exploratory analysis of income distributions at the disaggregated level.

This package provides implementation of methods for estimation of quantitative maps from Multi-Parameter Mapping (MPM) acquisitions including adaptive smoothing methods in the framework of the ESTATICS model. The smoothing method is described in Mohammadi et al. (2017). <doi:10.20347/WIAS.PREPRINT.2432>. Usage of the package is also described in Polzehl and Tabelow (2019), Magnetic Resonance Brain Imaging, Chapter 6, Springer, Use R! Series. <doi:10.1007/978-3-030-29184-6_6>.

Racket is a general-purpose programming language in the Scheme family, with a large set of libraries and a compiler based on Chez Scheme. Racket is also a platform for language-oriented programming, from small domain-specific languages to complete language implementations.

The main Racket distribution comes with many bundled packages, including the DrRacket IDE, libraries for GUI and web programming, and implementations of languages such as Typed Racket, R5RS and R6RS Scheme, Algol 60, and Datalog.

rifi analyses data from rifampicin time series created by microarray or RNAseq. rifi is a transcriptome data analysis tool for the holistic identification of transcription and decay associated processes. The decay constants and the delay of the onset of decay is fitted for each probe/bin. Subsequently, probes/bins of equal properties are combined into segments by dynamic programming, independent of a existing genome annotation. This allows to detect transcript segments of different stability or transcriptional events within one annotated gene. In addition to the classic decay constant/half-life analysis, rifi detects processing sites, transcription pausing sites, internal transcription start sites in operons, sites of partial transcription termination in operons, identifies areas of likely transcriptional interference by the collision mechanism and gives an estimate of the transcription velocity. All data are integrated to give an estimate of continous transcriptional units, i.e. operons. Comprehensive output tables and visualizations of the full genome result and the individual fits for all probes/bins are produced.

Augmented Regression with General Online data (ARGO) for accurate estimation of influenza epidemics in United States on national level, regional level and state level. It replicates the method introduced in paper Yang, S., Santillana, M. and Kou, S.C. (2015) <doi:10.1073/pnas.1515373112>; Ning, S., Yang, S. and Kou, S.C. (2019) <doi:10.1038/s41598-019-41559-6>; Yang, S., Ning, S. and Kou, S.C. (2021) <doi:10.1038/s41598-021-83084-5>.

Bindings to Google's C++ library Compact Language Detector 2 (see <https://github.com/cld2owners/cld2#readme> for more information). Probabilistically detects over 80 languages in plain text or HTML. For mixed-language input it returns the top three detected languages and their approximate proportion of the total classified text bytes (e.g. 80% English and 20% French out of 1000 bytes). There is also a cld3 package on CRAN which uses a neural network model instead.

This package provides measures to characterize the complexity of classification and regression problems based on aspects that quantify the linearity of the data, the presence of informative feature, the sparsity and dimensionality of the datasets. This package provides bug fixes, generalizations and implementations of many state of the art measures. The measures are described in the papers: Lorena et al. (2019) <doi:10.1145/3347711> and Lorena et al. (2018) <doi:10.1007/s10994-017-5681-1>.

This framework enables forecasting and extrapolating measures of conditional risk (e.g. of extreme or unprecedented events), including quantiles and exceedance probabilities, using extreme value statistics and flexible neural network architectures. It allows for capturing complex multivariate dependencies, including dependencies between observations, such as sequential dependence (time-series). The methodology was introduced in Pasche and Engelke (2024) <doi:10.1214/24-AOAS1907> (also available in preprint: Pasche and Engelke (2022) <doi:10.48550/arXiv.2208.07590>).

The algorithm assigns rareness/ outlierness score to every sample in voluminous datasets. The algorithm makes multiple estimations of the proximity between a pair of samples, in low-dimensional spaces. To compute proximity, FiRE uses Sketching, a variant of locality sensitive hashing. For more details: Jindal, A., Gupta, P., Jayadeva and Sengupta, D., 2018. Discovery of rare cells from voluminous single cell expression data. Nature Communications, 9(1), p.4719. <doi:10.1038/s41467-018-07234-6>.

This R package can be used to generate artificial data conditionally on pre-specified (simulated or user-defined) relationships between the variables and/or observations. Each observation is drawn from a multivariate Normal distribution where the mean vector and covariance matrix reflect the desired relationships. Outputs can be used to evaluate the performances of variable selection, graphical modelling, or clustering approaches by comparing the true and estimated structures (B Bodinier et al (2021) <arXiv:2106.02521>).

Retrieving regional plant checklists, species traits and distributions, and environmental data from the Global Inventory of Floras and Traits (GIFT). More information about the GIFT database can be found at <https://gift.uni-goettingen.de/about> and the map of available floras can be visualized at <https://gift.uni-goettingen.de/map>. The API and associated queries can be accessed according the following scheme: <https://gift.uni-goettingen.de/api/extended/index2.0.php?query=env_raster>.

Cross-validated eigenvalues are estimated by splitting a graph into two parts, the training and the test graph. The training graph is used to estimate eigenvectors, and the test graph is used to evaluate the correlation between the training eigenvectors and the eigenvectors of the test graph. The correlations follow a simple central limit theorem that can be used to estimate graph dimension via hypothesis testing, see Chen et al. (2021) <arXiv:2108.03336> for details.