This package provides functions are provided to fit temporal lag models to dynamic networks. The models are build on top of exponential random graph models (ERGM) framework. There are functions for simulating or forecasting networks for future time points. Abhirup Mallik & Zack W. Almquist (2019) Stable Multiple Time Step Simulation/Prediction From Lagged Dynamic Network Regression Models, Journal of Computational and Graphical Statistics, 28:4, 967-979, <DOI: 10.1080/10618600.2019.1594834>.

Collection of datasets as prepared by Profs. A.P. Gore, S.A. Paranjape, and M.B. Kulkarni of Department of Statistics, Poona University, India. With their permission, first letter of their names forms the name of this package, the package has been built by me and made available for the benefit of R users. This collection requires a rich class of models and can be a very useful building block for a beginner.

This package provides interpretable high-dimensional mean comparison methods (HMC). For example, users can apply these methods to assess the difference in gene expression between two treatment groups. It is not a gene-by-gene comparison. Instead, the methods focus on the interplay between features and identify those that are predictive of the group label. The tests are valid frequentist procedures and yield sparse estimates indicating which features contribute to the group differences.

An interface to the algorithms of Interpretable AI <https://www.interpretable.ai> from the R programming language. Interpretable AI provides various modules, including Optimal Trees for classification, regression, prescription and survival analysis, Optimal Imputation for missing data imputation and outlier detection, and Optimal Feature Selection for exact sparse regression. The iai package is an open-source project. The Interpretable AI software modules are proprietary products, but free academic and evaluation licenses are available.

Pattern Sequence Based Forecasting (PSF) takes univariate time series data as input and assist to forecast its future values. This algorithm forecasts the behavior of time series based on similarity of pattern sequences. Initially, clustering is done with the labeling of samples from database. The labels associated with samples are then used for forecasting the future behaviour of time series data. The further technical details and references regarding PSF are discussed in Vignette.

Derived from the work of Kruschke (2015, <ISBN:9780124058880>), the present package aims to provide a framework for conducting Bayesian analysis using Markov chain Monte Carlo (MCMC) sampling utilizing the Just Another Gibbs Sampler ('JAGS', Plummer, 2003, <https://mcmc-jags.sourceforge.io>). The initial version includes several modules for conducting Bayesian equivalents of chi-squared tests, analysis of variance (ANOVA), multiple (hierarchical) regression, softmax regression, and for fitting data (e.g., structural equation modeling).

This package provides a new methodology for linear regression with both curve response and curve regressors, which is described in Cho, Goude, Brossat and Yao (2013) <doi:10.1080/01621459.2012.722900> and (2015) <doi:10.1007/978-3-319-18732-7_3>. The key idea behind this methodology is dimension reduction based on a singular value decomposition in a Hilbert space, which reduces the curve regression problem to several scalar linear regression problems.

This package implements various decision support tools related to the Econometrics & Technometrics. Subroutines include correlation reliability test, Mahalanobis distance measure for outlier detection, combinatorial search (all possible subset regression), non-parametric efficiency analysis measures: DDF (directional distance function), DEA (data envelopment analysis), HDF (hyperbolic distance function), SBM (slack-based measure), and SF (shortage function), benchmarking, Malmquist productivity analysis, risk analysis, technology adoption model, new product target setting, network DEA, dynamic DEA, intertemporal budgeting, etc.

Computes the InterModel Vigorish ('IMV'), a metric for comparing the predictive accuracy of two models for binary outcomes. The IMV is derived from the expected value of a bettor using one model's predicted probabilities against those of a competing model, and is estimated via k-fold cross-validation. Methods are provided for generalized linear models, mixed-effects models ('lme4'), and item response theory models ('mirt'). See <doi:10.1371/journal.pone.0316491>.

Efficiently estimate shape parameters of periodic time series imagery with which a statistical seasonal trend analysis (STA) is subsequently performed. STA output can be exported in conventional raster formats. Methods to visualize STA output are also implemented as well as the calculation of additional basic statistics. STA is based on (R. Eastman, F. Sangermano, B. Ghimire, H. Zhu, H. Chen, N. Neeti, Y. Cai, E. Machado and S. Crema, 2009) <doi:10.1080/01431160902755338>.

Implementation of target-controlled infusion algorithms for compartmental pharmacokinetic and pharmacokinetic-pharmacodynamic models. Jacobs (1990) <doi:10.1109/10.43622>; Marsh et al. (1991) <doi:10.1093/bja/67.1.41>; Shafer and Gregg (1993) <doi:10.1007/BF01070999>; Schnider et al. (1998) <doi:10.1097/00000542-199805000-00006>; Abuhelwa, Foster, and Upton (2015) <doi:10.1016/j.vascn.2015.03.004>; Eleveld et al. (2018) <doi:10.1016/j.bja.2018.01.018>.

This package provides a new measure of similarity between a pair of mass spectrometry (MS) experiments, called truncated rank correlation (TRC). To provide a robust metric of similarity in noisy high-dimensional data, TRC uses truncated top ranks (or top m-ranks) for calculating correlation. Truncated rank correlation as a robust measure of test-retest reliability in mass spectrometry data. For more details see Lim et al. (2019) <doi:10.1515/sagmb-2018-0056>.

Variance function estimation for models proposed by W. Sadler in his variance function program ('VFP', www.aacb.asn.au/AACB/Resources/Variance-Function-Program). Here, the idea is to fit multiple variance functions to a data set and consequently assess which function reflects the relationship Var ~ Mean best. For in-vitro diagnostic ('IVD') assays modeling this relationship is of great importance when individual test-results are used for defining follow-up treatment of patients.

This package provides methods to calculate the expected value of information from a decision-analytic model. This includes the expected value of perfect information (EVPI), partial perfect information (EVPPI) and sample information (EVSI), and the expected net benefit of sampling (ENBS). A range of alternative computational methods are provided under the same user interface. See Heath et al. (2024) <doi:10.1201/9781003156109>, Jackson et al. (2022) <doi:10.1146/annurev-statistics-040120-010730>.

Alternating least squares is often used to resolve components contributing to data with a bilinear structure; the basic technique may be extended to alternating constrained least squares. This package provides an implementation of multivariate curve resolution alternating least squares (MCR-ALS).

Commonly applied constraints include unimodality, non-negativity, and normalization of components. Several data matrices may be decomposed simultaneously by assuming that one of the two matrices in the bilinear decomposition is shared between datasets.

This package provides tools for evaluating timely epidemic detection models within school absenteeism-based surveillance systems. Introduces the concept of alert time quality as an evaluation metric. Includes functions to simulate populations, epidemics, and alert metrics associated with epidemic spread using population census data. The methods are based on research published in Vanderkruk et al. (2023) <doi:10.1186/s12889-023-15747-z> and Ward et al. (2019) <doi:10.1186/s12889-019-7521-7>.

Using the Theory of Belief Functions for evidence calculus. Basic probability assignments, or mass functions, can be defined on the subsets of a set of possible values and combined. A mass function can be extended to a larger frame. Marginalization, i.e. reduction to a smaller frame can also be done. These features can be combined to analyze small belief networks and take into account situations where information cannot be satisfactorily described by probability distributions.

Enables the generation of Laplace factor models across diverse Laplace distributions and facilitates the application of Sparse Online Principal Component (SOPC), Incremental Principal Component (IPC), Perturbation Principal Component (PPC), Stochastic Approximation Principal Component (SAPC), Sparse Principal Component (SPC) and other PC methods and Farm Test methods to these models. Evaluates the efficacy of these methods within the context of Laplace factor models by scrutinizing parameter estimation accuracy, mean square error, and the degree of sparsity.

Eco-phylogenetic and community phylogenetic analyses. Keeps community ecological and phylogenetic data matched up and comparable using comparative.comm objects. Wrappers for common community phylogenetic indices ('pez.shape', pez.evenness', pez.dispersion', and pez.dissimilarity metrics). Implementation of Cavender-Bares (2004) correlation of phylogenetic and ecological matrices ('fingerprint.regression'). Phylogenetic Generalised Linear Mixed Models (PGLMMs; pglmm') following Ives & Helmus (2011) and Rafferty & Ives (2013). Simulation of null assemblages, traits, and phylogenies ('scape', sim.meta.comm').

Generates Skew Factor Models data and applies Sparse Online Principal Component (SOPC), Incremental Principal Component (IPC), Projected Principal Component (PPC), Perturbation Principal Component (PPC), Stochastic Approximation Principal Component (SAPC), Sparse Principal Component (SPC) and other PC methods to estimate model parameters. It includes capabilities for calculating mean squared error, relative error, and sparsity of the loading matrix.The philosophy of the package is described in Guo G. (2023) <doi:10.1007/s00180-022-01270-z>.

Implement the transformed additive Gaussian (TAG) process and the transformed approximately additive Gaussian (TAAG) process proposed in Lin and Joseph (2020) <DOI:10.1080/00401706.2019.1665592>. These functions can be used to model deterministic computer experiments, obtain predictions at new inputs, and quantify the uncertainty of the predictions. This research is supported by a U.S. National Science Foundation grant DMS-1712642 and a U.S. Army Research Office grant W911NF-17-1-0007.

Computes exact observation weights for the Kalman filter and smoother, following Koopman and Harvey (2003) <www.sciencedirect.com/science/article/pii/S0165188902000611>. The package provides tools for analyzing linear Gaussian state-space models, allowing users to quantify the contribution of individual observations to filtered and smoothed state estimates. These weights can be used for interpretation, decomposition, and diagnostic analysis in time series models, including applications such as dynamic factor models. See the README for examples.

Rapid realistic routing on multimodal transport networks (walk, bike, public transport and car) using R5', the Rapid Realistic Routing on Real-world and Reimagined networks engine <https://github.com/conveyal/r5>. The package allows users to generate detailed routing analysis or calculate travel time and monetary cost matrices using seamless parallel computing on top of the R5 Java machine. While R5 is developed by Conveyal, the package r5r is independently developed by a team at the Institute for Applied Economic Research (Ipea) with contributions from collaborators. Apart from the documentation in this package, users will find additional information on R5 documentation at <https://docs.conveyal.com/>. Although we try to keep new releases of r5r in synchrony with R5, the development of R5 follows Conveyal's independent update process. Hence, users should confirm the R5 version implied by the Conveyal user manual (see <https://docs.conveyal.com/changelog>) corresponds with the R5 version that r5r depends on. This version of r5r depends on R5 v7.1.

This algorithm conducts variable selection in the classification setting. It repeatedly subsamples variables and runs linear discriminant analysis (LDA) on the subsampled variables. Variables are scored based on the AUC and the t-statistics. Variables then enter a competition and the semi-finalist variables will be evaluated in a final round of LDA classification. The algorithm then outputs a list of variable selected. Qiao, Sun and Fan (2017) <http://people.math.binghamton.edu/qiao/swa.html>.