Linear or nonlinear cross-lagged panel model can be built from input data. Users can choose the appropriate method from three methods for constructing nonlinear cross lagged models. These three methods include polynomial regression, generalized additive model and generalized linear mixed model.In addition, a function for determining linear relationships is provided. Relevant knowledge of cross lagged models can be learned through the paper by Fredrik Falkenström (2024) <doi:10.1016/j.cpr.2024.102435> and the paper by A Gasparrini (2010) <doi:10.1002/sim.3940>.

The Large Language Model (LLM) represents a groundbreaking advancement in data science and programming, and also allows us to extend the world of R. A seamless interface for integrating the OpenAI Web APIs into R is provided in this package. This package leverages LLM-based AI techniques, enabling efficient knowledge discovery and data analysis (see OpenAI Web APIs details <https://openai.com/blog/openai-api>). The previous functions such as seamless translation and image generation have been moved to other packages deepRstudio and stableDiffusion4R'.

Evaluate diagnostic test performance using data from laboratory or diagnostic research. It supports both binary and continuous test variables. It allows users to compute key performance indicators and visualize Receiver Operating Characteristic (ROC) curves, determine optimal cut-off thresholds, display confusion matrix, and export publication-ready plot. It aims to facilitate the application of statistical methods in diagnostic test evaluation by healthcare professionals. The methodology used to compute the performance indicators follows the overview described by Habibzadeh (2025) <doi:10.11613/BM.2025.010101>. Thanks to shiny package.

The Australian Regulatory Guidelines for Prescription Medicines (ARGPM), guidance on "Stability testing for prescription medicines", recommends to predict the shelf life of chemically derived medicines from stability data by taking the worst case situation at batch release into account. Consequently, if a change over time is observed, a release limit needs to be specified. Finding a release limit and the associated shelf life is supported, as well as the standard approach that is recommended by guidance Q1E "Evaluation of stability data" from the International Council for Harmonisation (ICH).

This package provides a straightforward interface for accessing the IMF (International Monetary Fund) data JSON API, available at <https://data.imf.org/>. This package offers direct access to the primary API endpoints: Dataflow, DataStructure, and CompactData. And, it provides an intuitive interface for exploring available dimensions and attributes, as well as querying individual time-series datasets. Additionally, the package implements a rate limit on API calls to reduce the chances of exceeding service limits (limited to 10 calls every 5 seconds) and encountering response errors.

This package provides tools for reading and writing NIfTI-1.1 (NII) files, including optimized voxelwise read/write operations and a simplified method to write dataframes to NII. Specification of the NIfTI-1.1 format can be found here <https://nifti.nimh.nih.gov/nifti-1>. Scientific publication first using these tools Koscik TR, Man V, Jahn A, Lee CH, Cunningham WA (2020) <doi:10.1016/j.neuroimage.2020.116764> "Decomposing the neural pathways in a simple, value-based choice." Neuroimage, 214, 116764.

This package provides a robust approach for omics data integration and disease subtyping. PINSPlus is fast and supports the analysis of large datasets with hundreds of thousands of samples and features. The software automatically determines the optimal number of clusters and then partitions the samples in a way such that the results are robust against noise and data perturbation (Nguyen et al. (2019) <DOI: 10.1093/bioinformatics/bty1049>, Nguyen et al. (2017)<DOI: 10.1101/gr.215129.116>, Nguyen et al. (2021)<DOI: 10.3389/fonc.2021.725133>).

Allows the user to estimate a vector logistic smooth transition autoregressive model via maximum log-likelihood or nonlinear least squares. It further permits to test for linearity in the multivariate framework against a vector logistic smooth transition autoregressive model with a single transition variable. The estimation method is discussed in Terasvirta and Yang (2014, <doi:10.1108/S0731-9053(2013)0000031008>). Also, realized covariances can be constructed from stock market prices or returns, as explained in Andersen et al. (2001, <doi:10.1016/S0304-405X(01)00055-1>).

Preview spatial data as leaflet maps with minimal effort. smartmap is optimized for interactive use and distinguishes itself from similar packages because it does not need real spatial ('sp or sf') objects an input; instead, it tries to automatically coerce everything that looks like spatial data to sf objects or leaflet maps. It - for example - supports direct mapping of: a vector containing a single coordinate pair, a two column matrix, a data.frame with longitude and latitude columns, or the path or URL to a (possibly compressed) shapefile'.

Parametric survival regression models under the maximum likelihood approach via Stan'. Implemented regression models include accelerated failure time models, proportional hazards models, proportional odds models, accelerated hazard models, Yang and Prentice models, and extended hazard models. Available baseline survival distributions include exponential, Weibull, log-normal, log-logistic, gamma, generalized gamma, rayleigh, Gompertz and fatigue (Birnbaum-Saunders) distributions. References: Lawless (2002) <ISBN:9780471372158>; Bennett (1982) <doi:10.1002/sim.4780020223>; Chen and Wang(2000) <doi:10.1080/01621459.2000.10474236>; Demarqui and Mayrink (2021) <doi:10.1214/20-BJPS471>.

Characterisation of the extremal dependence structure of time series, avoiding pre-processing and filtering as done typically with peaks-over-threshold methods. It uses the conditional approach of Heffernan and Tawn (2004) <DOI:10.1111/j.1467-9868.2004.02050.x> which is very flexible in terms of extremal and asymptotic dependence structures, and Bayesian methods improve efficiency and allow for deriving measures of uncertainty. For example, the extremal index, related to the size of clusters in time, can be estimated and samples from its posterior distribution obtained.

This package provides additional data sets, methods and documentation to complement the vcd package for Visualizing Categorical Data and the gnm package for Generalized Nonlinear Models. In particular, vcdExtra extends mosaic, assoc and sieve plots from vcd to handle glm() and gnm() models and adds a 3D version in mosaic3d'. Additionally, methods are provided for comparing and visualizing lists of glm and loglm objects. This package is now a support package for the book, "Discrete Data Analysis with R" by Michael Friendly and David Meyer.

The jsonlite package provides a fast JSON parser and generator optimized for statistical data and the web. It offers flexible, robust, high performance tools for working with JSON in R and is particularly powerful for building pipelines and interacting with a web API. In addition to converting JSON data from/to R objects, jsonlite contains functions to stream, validate, and prettify JSON data. The unit tests included with the package verify that all edge cases are encoded and decoded consistently for use with dynamic data in systems and applications.

This is a package to provide infrastructure for managing package parameters. Parameters are easy to get in relevant functions within a package, and rrror is thrown if a parameter is missing. Developers are able to register parameters and set their default value in a config file that is part of the package in YAML format, and users are able to override parameters using their own YAML. Users get an exception when trying to override a parameter that was not registered, and can load multiple parameters to the current environment.

This package performs BTLLasso as described by Schauberger and Tutz (2019) <doi:10.18637/jss.v088.i09> and Schauberger and Tutz (2017) <doi:10.1177/1471082X17693086>. BTLLasso is a method to include different types of variables in paired comparison models and, therefore, to allow for heterogeneity between subjects. Variables can be subject-specific, object-specific and subject-object-specific and can have an influence on the attractiveness/strength of the objects. Suitable L1 penalty terms are used to cluster certain effects and to reduce the complexity of the models.

Useful tools for fitting, validating, and forecasting of practical convolution-closed time series models for low counts are provided. Marginal distributions of the data can be modelled via Poisson and Generalized Poisson innovations. Regression effects can be incorporated through time varying innovation rates. The models are described in Jung and Tremayne (2011) <doi:10.1111/j.1467-9892.2010.00697.x> and the model assessment tools are presented in Czado et al. (2009) <doi:10.1111/j.1541-0420.2009.01191.x> and, Tsay (1992) <doi:10.2307/2347612>.

This package provides a method to detect values poorly explained by a Gaussian linear model. The procedure is based on the maximum of the absolute value of the studentized residuals, which is a parameter-free statistic. This approach generalizes several procedures used to detect abnormal values during longitudinal monitoring of biological markers. For methodological details, see: Berthelot G., Saulière G., Dedecker J. (2025). "DEViaN-LM An R Package for Detecting Abnormal Values in the Gaussian Linear Model". HAL Id: hal-05230549. <https://hal.science/hal-05230549>.

Easy-to-use and efficient interface for Bayesian inference of complex panel (time series) data using dynamic multivariate panel models by Helske and Tikka (2024) <doi:10.1016/j.alcr.2024.100617>. The package supports joint modeling of multiple measurements per individual, time-varying and time-invariant effects, and a wide range of discrete and continuous distributions. Estimation of these dynamic multivariate panel models is carried out via Stan'. For an in-depth tutorial of the package, see (Tikka and Helske, 2024) <doi:10.48550/arXiv.2302.01607>.

The algorithm of semi-supervised learning based on finite Gaussian mixture models with a missing-data mechanism is designed for a fitting g-class Gaussian mixture model via maximum likelihood (ML). It is proposed to treat the labels of the unclassified features as missing-data and to introduce a framework for their missing as in the pioneering work of Rubin (1976) for missing in incomplete data analysis. This dependency in the missingness pattern can be leveraged to provide additional information about the optimal classifier as specified by Bayesâ rule.

Analyzing censored variables usually requires the use of optimization algorithms. This package provides an alternative algebraic approach to the task of determining the expected value of a random censored variable with a known censoring point. Likewise this approach allows for the determination of the censoring point if the expected value is known. These results are derived under the assumption that the variable follows an Epanechnikov kernel distribution with known mean and range prior to censoring. Statistical functions related to the uncensored Epanechnikov distribution are also provided by this package.

Gene information from Ensembl genome builds GRCh38.p14 and GRCh37.p13 to use with the topr package. The datasets were originally downloaded from <https://ftp.ensembl.org/pub/current/gtf/homo_sapiens/Homo_sapiens.GRCh38.111.gtf.gz> and <https://ftp.ensembl.org/pub/grch37/current/gtf/homo_sapiens/Homo_sapiens.GRCh37.87.gtf.gz> and converted into the format required by the topr package. See <https://github.com/totajuliusd/topr?tab=readme-ov-file#how-to-use-topr-with-other-species-than-human> to see the required format.

Routines for the estimation or simultaneous estimation and variable selection in several functional semiparametric models with scalar responses are provided. These models include the functional single-index model, the semi-functional partial linear model, and the semi-functional partial linear single-index model. Additionally, the package offers algorithms for handling scalar covariates with linear effects that originate from the discretization of a curve. This functionality is applicable in the context of the linear model, the multi-functional partial linear model, and the multi-functional partial linear single-index model.

Fits the (randomized drift) inverse Gaussian distribution to survival data. The model is described in Aalen OO, Borgan O, Gjessing HK. Survival and Event History Analysis. A Process Point of View. Springer, 2008. It is based on describing time to event as the barrier hitting time of a Wiener process, where drift towards the barrier has been randomized with a Gaussian distribution. The model allows covariates to influence starting values of the Wiener process and/or average drift towards a barrier, with a user-defined choice of link functions.

This package provides a collection of helper functions for multiple regression models fitted by lm(). Most of them are simple functions for simple tasks which can be done with coding, but may not be easy for occasional users of R. Most of the tasks addressed are those sometimes needed when using the manymome package (Cheung and Cheung, 2023, <doi:10.3758/s13428-023-02224-z>) and stdmod package (Cheung, Cheung, Lau, Hui, and Vong, 2022, <doi:10.1037/hea0001188>). However, they can also be used in other scenarios.