Multivariate Adaptive Regression Spline (MARS) based Artificial Neural Network (ANN) hybrid model is combined Machine learning hybrid approach which selects important variables using MARS and then fits ANN on the extracted important variables.

Sometimes data for analysis are obtained using more convenient or less expensive means yielding "surrogate" variables for what could be obtained more accurately, albeit with less convenience; or less conveniently or at more expense yielding "reference" variables, thought of as being measured without error. Analysis of the surrogate variables measured with error generally yields biased estimates when the objective is to make inference about the reference variables. Often it is thought that ignoring the measurement error in surrogate variables only biases effects toward the null hypothesis, but this need not be the case. Measurement errors may bias parameter estimates either toward or away from the null hypothesis. If one has a data set with surrogate variable data from the full sample, and also reference variable data from a randomly selected subsample, then one can assess the bias introduced by measurement error in parameter estimation, and use this information to derive improved estimates based upon all available data. Formulaically these estimates based upon the reference variables from the validation subsample combined with the surrogate variables from the whole sample can be interpreted as starting with the estimate from reference variables in the validation subsample, and "augmenting" this with additional information from the surrogate variables. This suggests the term "augmented" estimate. The meerva package calculates these augmented estimates in the regression setting when there is a randomly selected subsample with both surrogate and reference variables. Measurement errors may be differential or non-differential, in any or all predictors (simultaneously) as well as outcome. The augmented estimates derive, in part, from the multivariate correlation between regression model parameter estimates from the reference variables and the surrogate variables, both from the validation subset. Because the validation subsample is chosen at random any biases imposed by measurement error, whether non-differential or differential, are reflected in this correlation and these correlations can be used to derive estimates for the reference variables using data from the whole sample. The main functions in the package are meerva.fit which calculates estimates for a dataset, and meerva.sim.block which simulates multiple datasets as described by the user, and analyzes these datasets, storing the regression coefficient estimates for inspection. The augmented estimates, as well as how measurement error may arise in practice, is described in more detail by Kremers WK (2021) <arXiv:2106.14063> and is an extension of the works by Chen Y-H, Chen H. (2000) <doi:10.1111/1467-9868.00243>, Chen Y-H. (2002) <doi:10.1111/1467-9868.00324>, Wang X, Wang Q (2015) <doi:10.1016/j.jmva.2015.05.017> and Tong J, Huang J, Chubak J, et al. (2020) <doi:10.1093/jamia/ocz180>.

Many times, you will not find data for all dates. After first January, 2011 you may have next data on 20th January, 2011 and so on. Also available dates may have zero values. Try to gather all such kinds of data in different excel sheets of a single excel file. Every sheet will contain two columns (1st one is dates and second one is the data). After loading all the sheets into different elements of a list, using this you can fill the gaps for all the sheets and mark all the corresponding values as zeros. Here I am talking about daily data. Finally, it will combine all the filled results into one data frame (first column is date and other columns will be corresponding values of your sheets) and give one combined data frame. Number of columns in the data frame will be number of sheets plus one. Then imputation will be done. Daily to monthly and weekly conversion is also possible. More details can be found in Garai and others (2023) <doi:10.13140/RG.2.2.11977.42087>.

Relatively easy access is provided to 2023 version of the Maddison project data downloaded 2025-08-28. This project collates all the credible data on population and GDP for 169 countries, with some dating back to the year 1 of the current era. One function makes it easy to find the leaders for each year, allowing users to delete countries like OPEC with narrow economies to focus on technology leaders. Another function makes it easy to plot data for only selected countries or years. Another function makes it relatively easy to obtain references to the original sources, which must be cited per the copyright rules of the Maddison Project for different uses of their data.

Response Surface Designs (RSDs) involving factors not all at same levels are called Mixed Level RSDs (or Asymmetric RSDs). In many practical situations, RSDs with asymmetric levels will be more suitable as it explores more regions in the design space. (J.S. Mehta and M.N. Das (1968) <doi:10.2307/1267046>. "Asymmetric rotatable designs and orthogonal transformations").This package contains function named ATORDs_I() for generating asymmetric third order rotatable designs (ATORDs) based on third order designs given by Das and Narasimham (1962). Function ATORDs_II() generates asymmetric third order rotatable designs developed using t-design of unequal set sizes, which are smaller in size as compared to design generated by function ATORDs_I(). In general, third order rotatable designs can be classified into two classes viz., designs that are suitable for sequential experimentation and designs for non-sequential experimentation. The sequential experimentation approach involves conducting the trials step by step whereas, in the non-sequential experimentation approach, the entire runs are executed in one go (M. N. Das and V. Narasimham (1962) <doi:10.1214/AOMS/1177704374>. "Construction of Rotatable Designs through Balanced Incomplete Block Designs"). ATORDs_I() and ATORDs_II() functions generate non-sequential asymmetric third order designs. Function named SeqTORD() generates symmetric sequential third order design in blocks and also gives G-efficiency of the given design. Function named Asymseq() generates asymmetric sequential third order designs in blocks (M. Hemavathi, Eldho Varghese, Shashi Shekhar and Seema Jaggi (2020) <doi:10.1080/02664763.2020.1864817>. "Sequential asymmetric third order rotatable designs (SATORDs)"). In response surface design, situations may arise in which some of the factors are qualitative in nature (Jyoti Divecha and Bharat Tarapara (2017) <doi:10.1080/08982112.2016.1217338>. "Small, balanced, efficient, optimal, and near rotatable response surface designs for factorial experiments asymmetrical in some quantitative, qualitative factors"). The Function named QualRSD() generates second order design with qualitative factors along with their D-efficiency and G-efficiency. The function named RotatabilityQ() calculates a measure of rotatability (measure Q, 0 <= Q <= 1) given by Draper and Pukelshiem(1990) for given a design based on a second order model, (Norman R. Draper and Friedrich Pukelsheim(1990) <doi:10.1080/00401706.1990.10484635>. "Another look at rotatability").

Implement multiverse style analyses (Steegen S., Tuerlinckx F, Gelman A., Vanpaemal, W., 2016) <doi:10.1177/1745691616658637> to show the robustness of statistical inference. Multiverse analysis is a philosophy of statistical reporting where paper authors report the outcomes of many different statistical analyses in order to show how fragile or robust their findings are. The multiverse package (Sarma A., Kale A., Moon M., Taback N., Chevalier F., Hullman J., Kay M., 2021) <doi:10.31219/osf.io/yfbwm> allows users to concisely and flexibly implement multiverse-style analysis, which involve declaring alternate ways of performing an analysis step, in R and R Notebooks.

This package provides functions for diagnostic meta-analysis. Next to basic analysis and visualization the bivariate Model of Reitsma et al. (2005) that is equivalent to the HSROC of Rutter & Gatsonis (2001) can be fitted. A new approach based to diagnostic meta-analysis of Holling et al. (2012) is also available. Standard methods like summary, plot and so on are provided.

This package implements the moment-matching approximation for differences of non-standardized t-distributed random variables in both univariate and multivariate settings. The package provides density, distribution function, quantile function, and random generation for the approximated distributions of t-differences. The methodology establishes the univariate approximated distributions through the systematic matching of the first, second, and fourth moments, and extends it to multivariate cases, considering both scenarios of independent components and the more general multivariate t-distributions with arbitrary dependence structures. Methods build on the classical moment-matching approximation method (e.g., Casella and Berger (2024) <doi:10.1201/9781003456285>).

Model time series using mixture autoregressive (MAR) models. Implemented are frequentist (EM) and Bayesian methods for estimation, prediction and model evaluation. See Wong and Li (2002) <doi:10.1111/1467-9868.00222>, Boshnakov (2009) <doi:10.1016/j.spl.2009.04.009>), and the extensive references in the documentation.

An implementation of MLMC (Multi-Level Monte Carlo), Giles (2008) <doi:10.1287/opre.1070.0496>, Heinrich (1998) <doi:10.1006/jcom.1998.0471>, for R. This package builds on the original Matlab and C++ implementations by Mike Giles to provide a full MLMC driver and example level samplers. Multi-core parallel sampling of levels is provided built-in.

This package provides functions to support data cleaning, evaluation, and description, developed for integration with Maelstrom Research software tools. madshapR provides functions primarily to evaluate and manipulate datasets and data dictionaries in preparation for data harmonization with the package Rmonize and to facilitate integration and transfer between RStudio servers and secure Opal environments. madshapR functions can be used independently but are optimized in conjunction with â Rmonizeâ functions for streamlined and coherent harmonization processing.

Supports the generation of parallelogram, equilateral triangle, regular hexagon, isosceles trapezoid, Koch snowflake, hexaflake', Sierpinski triangle, Sierpinski carpet and Sierpinski trapezoid mazes via TurtleGraphics'. Mazes are generated by the recursive method: the domain is divided into sub-domains in which mazes are generated, then dividing lines with holes are drawn between them, see J. Buck, Recursive Division, <http://weblog.jamisbuck.org/2011/1/12/maze-generation-recursive-division-algorithm>.

Multivariate Surrogate Synchrony ('mvSUSY') estimates the synchrony within datasets that contain more than two time series. mvSUSY was developed from Surrogate Synchrony ('SUSY') with respect to implementing surrogate controls, and extends synchrony estimation to multivariate data. mvSUSY works as described in Meier & Tschacher (2021).

The ultimate goal is to support 2-2-1, 2-1-1, and 1-1-1 models for multilevel mediation, the option of a moderating variable for either the a, b, or both paths, and covariates. Currently the 1-1-1 model is supported and several options of random effects; the initial code for bootstrapping was evaluated in simulations by Falk, Vogel, Hammami, and MioÄ eviÄ (2024) <doi:10.3758/s13428-023-02079-4>. Support for Bayesian estimation using brms comprises ongoing work. Currently only continuous mediators and outcomes are supported. Factors for any predictors must be numerically represented.

This is a cross-platform linear model to SQL compiler. It generates SQL from linear and generalized linear models. Its interface consists of a single function, modelc(), which takes the output of lm() or glm() functions (or any object which has the same signature) and outputs a SQL character vector representing the predictions on the scale of the response variable as described in Dunn & Smith (2018) <doi:10.1007/978-1-4419-0118-7> and originating in Nelder & Wedderburn (1972) <doi:10.2307/2344614>. The resultant SQL can be included in a SELECT statement and returns output similar to that of the glm.predict() or lm.predict() predictions, assuming numeric types are represented in the database using sufficient precision. Currently log and identity link functions are supported.

This package contains several functions for statistical data analysis; e.g. for sample size and power calculations, computation of confidence intervals and tests, and generation of similarity matrices.

The main objective of this package is to support the definition of Moodle elements taking advantage of the power that R offers. In this first version, it allows the definition of quizzes to be included in the question bank.

Compose generic monadic function pipelines with %>>% and %>-% based on implementing the S7 generics fmap() and bind(). Methods are provided for the built-in list type and the maybe class from the maybe package. The concepts are modelled directly after the Monad typeclass in Haskell, but adapted for idiomatic use in R.

Modern model-based geostatistics for point-referenced data. This package provides a simple interface to run spatial machine learning models and geostatistical models that estimate a continuous (raster) surface from point-referenced outcomes and, optionally, a set of raster covariates. The package also includes functions to summarize raster outcomes by (polygon) region while preserving uncertainty.

Framework for the Item Response Theory analysis of dichotomous and ordinal polytomous outcomes under the assumption of multidimensionality and discreteness of the latent traits. The fitting algorithms allow for missing responses and for different item parameterizations and are based on the Expectation-Maximization paradigm. Individual covariates affecting the class weights may be included in the new version (since 2.1).

Extends the mlr3 package with a connector to the package batchtools'. This allows to run large-scale benchmark experiments on scheduled high-performance computing clusters.

Model fitting and simulation for Gaussian and logistic inner product MultiNeSS models for multiplex networks. The package implements a convex fitting algorithm with fully adaptive parameter tuning, including options for edge cross-validation. For more details see MacDonald et al. (2020).

This package implements multivariate Fay-Herriot models for small area estimation. It uses empirical best linear unbiased prediction (EBLUP) estimator. Multivariate models consider the correlation of several target variables and borrow strength from auxiliary variables to improve the effectiveness of a domain sample size. Models which accommodated by this package are univariate model with several target variables (model 0), multivariate model (model 1), autoregressive multivariate model (model 2), and heteroscedastic autoregressive multivariate model (model 3). Functions provide EBLUP estimators and mean squared error (MSE) estimator for each model. These models were developed by Roberto Benavent and Domingo Morales (2015) <doi:10.1016/j.csda.2015.07.013>.

An implementation of the iterative proportional fitting (IPFP), maximum likelihood, minimum chi-square and weighted least squares procedures for updating a N-dimensional array with respect to given target marginal distributions (which, in turn can be multidimensional). The package also provides an application of the IPFP to simulate multivariate Bernoulli distributions.