This package provides functions for implementing the Generalized Bayesian Optimal Phase II (G-BOP2) design using various Particle Swarm Optimization (PSO) algorithms, including: - PSO-Default, based on Kennedy and Eberhart (1995) <doi:10.1109/ICNN.1995.488968>, "Particle Swarm Optimization"; - PSO-Quantum, based on Sun, Xu, and Feng (2004) <doi:10.1109/ICCIS.2004.1460396>, "A Global Search Strategy of Quantum-Behaved Particle Swarm Optimization"; - PSO-Dexp, based on Stehlà k et al. (2024) <doi:10.1016/j.asoc.2024.111913>, "A Double Exponential Particle Swarm Optimization with Non-Uniform Variates as Stochastic Tuning and Guaranteed Convergence to a Global Optimum with Sample Applications to Finding Optimal Exact Designs in Biostatistics"; - and PSO-GO.

Combining genomic prediction with Monte Carlo simulation, three different strategies are implemented to select parental lines for multiple traits in plant breeding. The selection strategies include (i) GEBV-O considers only genomic estimated breeding values (GEBVs) of the candidate individuals; (ii) GD-O considers only genomic diversity (GD) of the candidate individuals; and (iii) GEBV-GD considers both GEBV and GD. The above method can be seen in Chung PY, Liao CT (2020) <doi:10.1371/journal.pone.0243159>. Multi-trait genomic best linear unbiased prediction (MT-GBLUP) model is used to simultaneously estimate GEBVs of the target traits, and then a selection index is adopted to evaluate the composite performance of an individual.

Hidden Markov Model (HMM) based on symmetric lambda distribution framework is implemented for the study of return time-series in the financial market. Major features in the S&P500 index, such as regime identification, volatility clustering, and anti-correlation between return and volatility, can be extracted from HMM cleanly. Univariate symmetric lambda distribution is essentially a location-scale family of exponential power distribution. Such distribution is suitable for describing highly leptokurtic time series obtained from the financial market. It provides a theoretically solid foundation to explore such data where the normal distribution is not adequate. The HMM implementation follows closely the book: "Hidden Markov Models for Time Series", by Zucchini, MacDonald, Langrock (2016).

Microbial growth is often measured by growth curves i.e. a table of population sizes and times of measurements. This package allows to use such growth curve data to determine the duration of "microbial lag phase" i.e. the time needed for microbes to restart divisions. It implements the most commonly used methods to calculate the lag duration, these methods are discussed and described in Opalek et.al. 2022. Citation: Smug, B. J., Opalek, M., Necki, M., & Wloch-Salamon, D. (2024). Microbial lag calculator: A shiny-based application and an R package for calculating the duration of microbial lag phase. Methods in Ecology and Evolution, 15, 301â 307 <doi:10.1111/2041-210X.14269>.

Generates chronological and ordered p-plots for data vectors or vectors of p-values. The p-plot visualizes the evolution of the p-value of a significance test across the sampled data. It allows for assessing the consistency of the observed effects, for detecting the presence of potential moderator variables, and for estimating the influence of outlier values on the observed results. For non-significant findings, it can diagnose patterns indicative of underpowered study designs. The p-plot can thus either back the binary accept-vs-reject decision of common null-hypothesis significance tests, or it can qualify this decision and stimulate additional empirical work to arrive at more robust and replicable statistical inferences.

COCOA is a method for understanding epigenetic variation among samples. COCOA can be used with epigenetic data that includes genomic coordinates and an epigenetic signal, such as DNA methylation and chromatin accessibility data. To describe the method on a high level, COCOA quantifies inter-sample variation with either a supervised or unsupervised technique then uses a database of "region sets" to annotate the variation among samples. A region set is a set of genomic regions that share a biological annotation, for instance transcription factor (TF) binding regions, histone modification regions, or open chromatin regions. COCOA can identify region sets that are associated with epigenetic variation between samples and increase understanding of variation in your data.

Perform hierarchical Bayesian Aldrich-McKelvey scaling using Hamiltonian Monte Carlo via Stan'. Aldrich-McKelvey ('AM') scaling is a method for estimating the latent positions of survey respondents and external objects on a common scale using positional survey data. The hierarchical versions of the Bayesian AM model included in this package outperform other versions both in terms of yielding meaningful posterior distributions for respondent positions and in terms of recovering true respondent positions in simulations. The package contains functions for preparing data, fitting models, extracting estimates, plotting key results, and comparing models using cross-validation. The original version of the default model is described in Bølstad (2024) <doi:10.1017/pan.2023.18>.

Optimal k Nearest Neighbours Ensemble is an ensemble of base k nearest neighbour models each constructed on a bootstrap sample with a random subset of features. k closest observations are identified for a test point "x" (say), in each base k nearest neighbour model to fit a stepwise regression to predict the output value of "x". The final predicted value of "x" is the mean of estimates given by all the models. The implemented model takes training and test datasets and trains the model on training data to predict the test data. Ali, A., Hamraz, M., Kumam, P., Khan, D.M., Khalil, U., Sulaiman, M. and Khan, Z. (2020) <DOI:10.1109/ACCESS.2020.3010099>.

This package provides functions for pooling/combining the results (i.e., p-values) from (dependent) hypothesis tests. Included are Fisher's method, Stouffer's method, the inverse chi-square method, the Bonferroni method, Tippett's method, and the binomial test. Each method can be adjusted based on an estimate of the effective number of tests or using empirically derived null distribution using pseudo replicates. For Fisher's, Stouffer's, and the inverse chi-square method, direct generalizations based on multivariate theory are also available (leading to Brown's method, Strube's method, and the generalized inverse chi-square method). An introduction can be found in Cinar and Viechtbauer (2022) <doi:10.18637/jss.v101.i01>.

This package provides functions for simulating Markov chains using the Barker proposal to compute Markov chain Monte Carlo (MCMC) estimates of expectations with respect to a target distribution on a real-valued vector space. The Barker proposal, described in Livingstone and Zanella (2022) <doi:10.1111/rssb.12482>, is a gradient-based MCMC algorithm inspired by the Barker accept-reject rule. It combines the robustness of simpler MCMC schemes, such as random-walk Metropolis, with the efficiency of gradient-based methods, such as the Metropolis adjusted Langevin algorithm. The key function provided by the package is sample_chain(), which allows sampling a Markov chain with a specified target distribution as its stationary distribution. The chain is sampled by generating proposals and accepting or rejecting them using a Metropolis-Hasting acceptance rule. During an initial warm-up stage, the parameters of the proposal distribution can be adapted, with adapters available to both: tune the scale of the proposals by coercing the average acceptance rate to a target value; tune the shape of the proposals to match covariance estimates under the target distribution. As well as the default Barker proposal, the package also provides implementations of alternative proposal distributions, such as (Gaussian) random walk and Langevin proposals. Optionally, if BridgeStan's R interface <https://roualdes.us/bridgestan/latest/languages/r.html>, available on GitHub <https://github.com/roualdes/bridgestan>, is installed, then BridgeStan can be used to specify the target distribution to sample from.

Set of tools for fitting the additive partial linear models with symmetric autoregressive errors of order p, or APLMS-AR(p). This setup enables the modeling of a time series response variable using linear and nonlinear structures of a set of explanatory variables, with nonparametric components approximated by natural cubic splines or P-splines. It also accounts for autoregressive error terms with distributions that have lighter or heavier tails than the normal distribution. The package includes various error distributions, such as normal, generalized normal, Student's t, generalized Student's t, power-exponential, and Cauchy distributions. Chou-Chen, S.W., Oliveira, R.A., Raicher, I., Gilberto A. Paula (2024) <doi:10.1007/s00362-024-01590-w>.

This package provides methods for (auto)covariance/correlation function estimation in change point regression with stationary errors circumventing the pre-estimation of the underlying signal of the observations. Generic, first-order, (m+1)-gapped, difference-based autocovariance function estimator is based on M. Levine and I. Tecuapetla-Gómez (2023) <doi:10.48550/arXiv.1905.04578>. Bias-reducing, second-order, (m+1)-gapped, difference-based estimator is based on I. Tecuapetla-Gómez and A. Munk (2017) <doi:10.1111/sjos.12256>. Robust autocovariance estimator for change point regression with autoregressive errors is based on S. Chakar et al. (2017) <doi:10.3150/15-BEJ782>. It also includes a general projection-based method for covariance matrix estimation.

In the USA, companies file different forms with the U.S. Securities and Exchange Commission (SEC) through EDGAR (Electronic Data Gathering, Analysis, and Retrieval system). The EDGAR database automated system collects all the different necessary filings and makes it publicly available. This package facilitates retrieving, storing, searching, and parsing of all the available filings on the EDGAR server. It downloads filings from SEC server in bulk with a single query. Additionally, it provides various useful functions: extracts 8-K triggering events, extract "Business (Item 1)" and "Management's Discussion and Analysis(Item 7)" sections of annual statements, searches filings for desired keywords, provides sentiment measures, parses filing header information, and provides HTML view of SEC filings.

This package provides a tool for computing probabilities and other quantities that are relevant in selecting performance criteria for discrete trial training. The main function, miebl(), computes Bayesian and frequentist probabilities and bounds for each of n possible performance criterion choices when attempting to determine a student's true mastery level by counting their number of successful attempts at displaying learning among n trials. The reporting function miebl_re() takes output from miebl() and prepares it into a brief report for a specific criterion. miebl_cp() combines 2 to 5 distributions of true mastery level given performance criterion in one plot for comparison. Ramos (2025) <doi:10.1007/s40617-025-01058-9>.

The classical two-sample t-test works well for the normally distributed data or data with large sample size. The tcfu() and tt() tests implemented in this package provide better type-I-error control with more accurate power when testing the equality of two-sample means for skewed populations having unequal variances. These tests are especially useful when the sample sizes are moderate. The tcfu() uses the Cornish-Fisher expansion to achieve a better approximation to the true percentiles. The tt() provides transformations of the Welch's t-statistic so that the sampling distribution become more symmetric. For more technical details, please refer to Zhang (2019) <http://hdl.handle.net/2097/40235>.

Consolidates and calculates different sets of time-series features from multiple R and Python packages including Rcatch22 Henderson, T. (2021) <doi:10.5281/zenodo.5546815>, feasts O'Hara-Wild, M., Hyndman, R., and Wang, E. (2021) <https://CRAN.R-project.org/package=feasts>, tsfeatures Hyndman, R., Kang, Y., Montero-Manso, P., Talagala, T., Wang, E., Yang, Y., and O'Hara-Wild, M. (2020) <https://CRAN.R-project.org/package=tsfeatures>, tsfresh Christ, M., Braun, N., Neuffer, J., and Kempa-Liehr A.W. (2018) <doi:10.1016/j.neucom.2018.03.067>, TSFEL Barandas, M., et al. (2020) <doi:10.1016/j.softx.2020.100456>, and Kats Facebook Infrastructure Data Science (2021) <https://facebookresearch.github.io/Kats/>.

Computes Chatterjee's non-parametric correlation coefficient for time series data. It extends the original metric to time series analysis by providing the univariate Xi-Autocorrelation Function (Xi-ACF), directional Xi-Cross-Correlation Function (Xi-CCF), and multivariate network evaluation matrices. The package allows users to test for non-linear dependence using Iterative Amplitude Adjusted Fourier Transform (IAAFT) and Multivariate IAAFT (MIAAFT) surrogate data with strict Family-Wise Error Rate ('FWER') control via Max-statistic approaches. Methodologies are based on Chatterjee (2021) <doi:10.1080/01621459.2020.1758115>, surrogate data testing methods by Schreiber and Schmitz (1996) <doi:10.1103/PhysRevLett.77.635>, and local structural identification by Watanabe (2026) <doi:10.2139/ssrn.6829431>.

The skewr package is a tool for visualizing the output of the Illumina Human Methylation 450k BeadChip to aid in quality control. It creates a panel of nine plots. Six of the plots represent the density of either the methylated intensity or the unmethylated intensity given by one of three subsets of the 485,577 total probes. These subsets include Type I-red, Type I-green, and Type II.The remaining three distributions give the density of the Beta-values for these same three subsets. Each of the nine plots optionally displays the distributions of the "rs" SNP probes and the probes associated with imprinted genes as series of tick marks located above the x-axis.

The developed function is a comprehensive tool for the analysis of India Meteorological Department (IMD) NetCDF rainfall data. Specifically designed to process high-resolution daily gridded rainfall datasets. It provides four key functions to process IMD NetCDF rainfall data and create rasters for various temporal scales, including annual, seasonal, monthly, and weekly rainfall. For method details see, Malik, A. (2019).<DOI:10.1007/s12517-019-4454-5>. It supports different aggregation methods, such as sum, min, max, mean, and standard deviation. These functions are designed for spatio-temporal analysis of rainfall patterns, trend analysis,geostatistical modeling of rainfall variability, identifying rainfall anomalies and extreme events and can be an input for hydrological and agricultural models.

This package implements the Capability-Ecological Developmental Model (CEDM) for longitudinal and multilevel data. The package supports estimation and interpretation of models examining how socioeconomic status (SES), health indicators, and contextual factors jointly relate to academic outcomes. Functionality includes: (1) classification of ecological capability regimes (amplifying, neutral, compensatory); (2) estimation of moderated multilevel models with higher-order interaction terms; (3) causal mediation analysis using doubly robust estimation; (4) random-effects within-between (REWB) decomposition; (5) nonlinear moderation using restricted cubic splines; (6) clustering of longitudinal health trajectories; and (7) sensitivity analysis using the impact threshold for a confounding variable (ITCV) and robustness-to-replacement (RIR) measures. The package is designed for use with general longitudinal multilevel datasets.

This package provides a d-statistic tests the null hypothesis of no treatment effect in a matched, nonrandomized study of the effects caused by treatments. A d-statistic focuses on subsets of matched pairs that demonstrate insensitivity to unmeasured bias in such an observational study, correcting for double-use of the data by conditional inference. This conditional inference can, in favorable circumstances, substantially increase the power of a sensitivity analysis (Rosenbaum (2010) <doi:10.1007/978-1-4419-1213-8_14>). There are two examples, one concerning unemployment from Lalive et al. (2006) <doi:10.1111/j.1467-937X.2006.00406.x>, the other concerning smoking and periodontal disease from Rosenbaum (2017) <doi:10.1214/17-STS621>.

The functions compute the double-entry intraclass correlation, which is an index of profile similarity (Furr, 2010; McCrae, 2008). The double-entry intraclass correlation is a more precise index of the agreement of two empirically observed profiles than the often-used intraclass correlation (McCrae, 2008). Profiles comprising correlations are automatically transformed according to the Fisher z-transformation before the double-entry intraclass correlation is calculated. If the profiles comprise scores such as sum scores from various personality scales, it is recommended to standardize each individual score prior to computation of the double-entry intraclass correlation (McCrae, 2008). See Furr (2010) <doi:10.1080/00223890903379134> or McCrae (2008) <doi:10.1080/00223890701845104> for details.

Map image classification efficacy (MICE) adjusts the accuracy rate relative to a random classification baseline (Shao et al. (2021)<doi:10.1109/ACCESS.2021.3116526> and Tang et al. (2024)<doi:10.1109/TGRS.2024.3446950>). Only the proportions from the reference labels are considered, as opposed to the proportions from the reference and predictions, as is the case for the Kappa statistic. This package offers means to calculate MICE and adjusted versions of class-level user's accuracy (i.e., precision) and producer's accuracy (i.e., recall) and F1-scores. Class-level metrics are aggregated using macro-averaging. Functions are also made available to estimate confidence intervals using bootstrapping and statistically compare two classification results.

This package implements various methods for eliciting a probability distribution for a single parameter from an expert or a group of experts. The expert provides a small number of probability judgements, corresponding to points on his or her cumulative distribution function. A range of parametric distributions can then be fitted and displayed, with feedback provided in the form of fitted probabilities and percentiles. For multiple experts, a weighted linear pool can be calculated. Also includes functions for eliciting beliefs about population distributions; eliciting multivariate distributions using a Gaussian copula; eliciting a Dirichlet distribution; eliciting distributions for variance parameters in a random effects meta-analysis model; survival extrapolation. R Shiny apps for most of the methods are included.