This package provides methods for detecting influential subjects in longitudinal data, particularly when observations are collected at irregular time points. The package identifies subjects whose response trajectories deviate substantially from population-level patterns, helping to diagnose anomalies and undue influence on model estimates.

Regression models for interval censored data. Currently supports Cox-PH, proportional odds, and accelerated failure time models. Allows for semi and fully parametric models (parametric only for accelerated failure time models) and Bayesian parametric models. Includes functions for easy visual diagnostics of model fits and imputation of censored data.

Network functionalities specialized for data generated from input-output tables.

The Dynamic Time Warping (DTW) distance measure for time series allows non-linear alignments of time series to match similar patterns in time series of different lengths and or different speeds. IncDTW is characterized by (1) the incremental calculation of DTW (reduces runtime complexity to a linear level for updating the DTW distance) - especially for life data streams or subsequence matching, (2) the vector based implementation of DTW which is faster because no matrices are allocated (reduces the space complexity from a quadratic to a linear level in the number of observations) - for all runtime intensive DTW computations, (3) the subsequence matching algorithm runDTW, that efficiently finds the k-NN to a query pattern in a long time series, and (4) C++ in the heart. For details about DTW see the original paper "Dynamic programming algorithm optimization for spoken word recognition" by Sakoe and Chiba (1978) <DOI:10.1109/TASSP.1978.1163055>. For details about this package, Dynamic Time Warping and Incremental Dynamic Time Warping please see "IncDTW: An R Package for Incremental Calculation of Dynamic Time Warping" by Leodolter et al. (2021) <doi:10.18637/jss.v099.i09>.

This package provides a collection of Item Response Theory (IRT) and Computerized Adaptive Testing (CAT) functions that are used in psychometrics.

Creation of tables of summary statistics or counts for clinical data (for TLFs'). These tables can be exported as in-text table (with the flextable package) for a Clinical Study Report (Word format) or a topline presentation (PowerPoint format), or as interactive table (with the DT package) to an html document for clinical data review.

Develops a General Equilibrium (GE) Model, which estimates key variables such as wages, the number of residents and workers, the prices of the floor space, and its distribution between commercial and residential use, as in Ahlfeldt et al., (2015) <doi:10.3982/ECTA10876>. By doing so, the model allows understanding the economic influence of different urban policies.

As a sequel to iNEXT', the iNEXT.beta3D package provides functions to compute standardized taxonomic, phylogenetic, and functional diversity (3D) estimates with a common sample size (for alpha and gamma diversity) or sample coverage (for alpha, beta, gamma diversity as well as dissimilarity or turnover indices). Hill numbers and their generalizations are used to quantify 3D and to make multiplicative decomposition (gamma = alpha x beta). The package also features size- and coverage-based rarefaction and extrapolation sampling curves to facilitate rigorous comparison of beta diversity across datasets. See Chao et al. (2023) <doi:10.1002/ecm.1588> for more details.

Estimation of joint models for multivariate longitudinal markers (with various distributions available) and survival outcomes (possibly accounting for competing risks) with Integrated Nested Laplace Approximations (INLA). The flexible and user friendly function joint() facilitates the use of the fast and reliable inference technique implemented in the INLA package for joint modeling. More details are given in the help page of the joint() function (accessible via ?joint in the R console) and the vignette associated to the joint() function (accessible via vignette("INLAjoint") in the R console).

This package implements inequality constrained inference. This includes parameter estimation in normal (linear) models under linear equality and inequality constraints, as well as normal likelihood ratio tests involving inequality-constrained hypotheses. For inequality-constrained linear models, averaging over R-squared for different orderings of regressors is also included.

Introductory statistics methods to accompany "Investigating Statistical Concepts, Applications, and Methods" (ISCAM) by Beth Chance & Allan Rossman (2024) <https://rossmanchance.com/iscam4/>. Tools to introduce statistical concepts with a focus on simulation approaches. Functions are verbose, designed to provide ample output for students to understand what each function does. Additionally, most functions are accompanied with plots. The package is designed to be used in an educational setting alongside the ISCAM textbook.

The general workflow of most imputation methods is quite similar. The aim of this package is to provide parts of this general workflow to make the implementation of imputation methods easier. The heart of an imputation method is normally the used model. These models can be defined using the parsnip package or customized specifications. The rest of an imputation method are more technical specification e.g. which columns and rows should be used for imputation and in which order. These technical specifications can be set inside the imputation functions.

Fit a predictive model using iteratively reweighted boosting (IRBoost) to minimize robust loss functions within the CC-family (concave-convex). This constitutes an application of iteratively reweighted convex optimization (IRCO), where convex optimization is performed using the functional descent boosting algorithm. IRBoost assigns weights to facilitate outlier identification. Applications include robust generalized linear models and robust accelerated failure time models. Wang (2025) <doi:10.6339/24-JDS1138>.

Using embedded sdmx queries, get the data of more than 150 000 insee series from bdm macroeconomic database.

The itdr() routine allows for the estimation of sufficient dimension reduction subspaces in univariate regression such as the central mean subspace or central subspace in regression. This is achieved using Fourier transformation methods proposed by Zhu and Zeng (2006) <doi:10.1198/016214506000000140>, convolution transformation methods proposed by Zeng and Zhu (2010) <doi:10.1016/j.jmva.2009.08.004>, and iterative Hessian transformation methods proposed by Cook and Li (2002) <doi:10.1214/aos/1021379861>. Additionally, mitdr() function provides optimal estimators for sufficient dimension reduction subspaces in multivariate regression by optimizing a discrepancy function using a Fourier transform approach proposed by Weng and Yin (2022) <doi:10.5705/ss.202020.0312>, and selects the sufficient variables using Fourier transform sparse inverse regression estimators proposed by Weng (2022) <doi:10.1016/j.csda.2021.107380>.

The matrix factor model has drawn growing attention for its advantage in achieving two-directional dimension reduction simultaneously for matrix-structured observations. In contrast to the Principal Component Analysis (PCA)-based methods, we propose a simple Iterative Alternating Least Squares (IALS) algorithm for matrix factor model, see the details in He et al. (2023) <arXiv:2301.00360>.

Single Layer Feed-forward Neural networks (SLFNs) have many applications in various fields of statistical modelling, especially for time-series forecasting. However, there are some major disadvantages of training such networks via the widely accepted gradient-based backpropagation algorithm, such as convergence to local minima, dependencies on learning rate and large training time. These concerns were addressed by Huang et al. (2006) <doi:10.1016/j.neucom.2005.12.126>, wherein they introduced the Extreme Learning Machine (ELM), an extremely fast learning algorithm for SLFNs which randomly chooses the weights connecting input and hidden nodes and analytically determines the output weights of SLFNs. It shows good generalized performance, but is still subject to a high degree of randomness. To mitigate this issue, this package uses a dimensionality reduction technique given in Hyvarinen (1999) <doi:10.1109/72.761722>, namely, the Independent Component Analysis (ICA) to determine the input-hidden connections and thus, remove any sort of randomness from the algorithm. This leads to a robust, fast and stable ELM model. Using functions within this package, the proposed model can also be compared with an existing alternative based on the Principal Component Analysis (PCA) algorithm given by Pearson (1901) <doi:10.1080/14786440109462720>, i.e., the PCA based ELM model given by Castano et al. (2013) <doi:10.1007/s11063-012-9253-x>, from which the implemented ICA based algorithm is greatly inspired.

This package provides a collection of wrapper functions for common variable and dataset manipulation workflows primarily used by iNZight', a graphical user interface providing easy exploration and visualisation of data for students of statistics, available in both desktop and online versions. Additionally, many of the functions return the tidyverse code used to obtain the result in an effort to bridge the gap between GUI and coding.

Reads the output of the PerkinElmer InForm software <http://www.perkinelmer.com/product/inform-cell-analysis-one-seat-cls135781>. In addition to cell-density count, it can derive statistics of intercellular spatial distance for each cell-type.

Carries out integrative clustering analysis using multiple types of genomic dataset using integrative Non-negative Matrix factorization.

Multi-data type subtyping, which is data type agnostic and accepts missing data. Subtyping is performed using intermediary assessments created with autoencoders and similarity calculations. See Fox et al. (2024) <doi:10.1016/j.crmeth.2024.100884> for details.

This package implements continuous-time hidden Markov models (HMMs) to infer identity-by-descent (IBD) segments shared by two individuals from their single-nucleotide polymorphism (SNP) genotypes. Provides posterior probabilities at each marker (forward-backward algorithm), prediction of IBD segments (Viterbi algorithm), and functions for visualising results. Supports both autosomal data and X-chromosomal data.

Package for training interpretable machine learning models. Historically, the most interpretable machine learning models were not very accurate, and the most accurate models were not very interpretable. Microsoft Research has developed an algorithm called the Explainable Boosting Machine (EBM) which has both high accuracy and interpretable characteristics. EBM uses machine learning techniques like bagging and boosting to breathe new life into traditional GAMs (Generalized Additive Models). This makes them as accurate as random forests and gradient boosted trees, and also enhances their intelligibility and editability. Details on the EBM algorithm can be found in the paper by Rich Caruana, Yin Lou, Johannes Gehrke, Paul Koch, Marc Sturm, and Noemie Elhadad (2015, <doi:10.1145/2783258.2788613>).

Finds optimal designs for nonlinear models using a metaheuristic algorithm called Imperialist Competitive Algorithm (ICA). See, for details, Masoudi et al. (2022) <doi:10.32614/RJ-2022-043>, Masoudi et al. (2017) <doi:10.1016/j.csda.2016.06.014> and Masoudi et al. (2019) <doi:10.1080/10618600.2019.1601097>.