A Variational Bayesian algorithm for high-dimensional multi-source heterogeneous linear models. More details have been written up in a paper submitted to the journal Statistics in Medicine, and the details of variational Bayesian methods can be found in Ray and Szabo (2021) &lt;<a href="https://doi.org/10.1080%2F01621459.2020.1847121" target="_top">doi:10.1080/01621459.2020.1847121</a>&gt;. It simultaneously performs parameter estimation and variable selection. The algorithm supports two model settings: (1) local models, where variable selection is only applied to homogeneous coefficients, and (2) global models, where variable selection is also performed on heterogeneous coefficients. Two forms of Spike-and-Slab priors are available: the Laplace distribution and the Gaussian distribution as the Slab component.

Detects values imported from spreadsheets that were auto-converted to Excel date serials and reconstructs the originally intended day.month decimals (for example, '30.3' that Excel displayed as '30/03/2025'). The functions work in a vectorized manner, preserve non-serial values, and support both the 1900 and 1904 date systems.

Provides functions to construct two-phase design layouts, compute treatment- and block-incidence matrices, derive C-matrices for residual, direct, and interaction effects, and calculate the efficiency factor for two-phase experimental designs with factorial treatment structure.

Calculate geolocations by light using template matching. The routine uses a calibration free optimization of a sky illuminance model to determine locations robustly using a template matching approach, as described by Ekstrom (2004) &lt;<a href="https://nipr.repo.nii.ac.jp/records/2496" target="_top">https://nipr.repo.nii.ac.jp/records/2496</a>&gt;, and behaviourly informed constraints (step-selection).

Enable users to measure and record the execution time of pipe operations (using |&gt;) with optional logging to dataframes and output to the console.

Creates a spectroscopy guideline with a highly accurate prediction model for soil properties using machine learning or deep learning algorithms such as LASSO, Random Forest, Cubist, etc., and decide which algorithm generates the best model for different soil types.

https://cran.r-project.org/package=FuzzySpec

Simulate a virtual population of subjects that has demographic distributions (height, weight, and BMI) and correlations (height and weight), by sex and age, which mimic those reported in real-world anthropometric growth charts (CDC, WHO, or Fenton).

Collection of tools for the analysis of the resilience of dynamic networks. Created as a classroom project.