# Alan Miller

#### 5 packages on CRAN

Implementation of various default Bayes factors for testing statistical hypotheses. The package is intended for applied quantitative researchers in the social and behavioral sciences, medical research, and related fields. The Bayes factor tests can be executed for statistical models such as univariate and multivariate normal linear models, generalized linear models, special cases of linear mixed models, survival models, relational event models. Parameters that can be tested are location parameters (e.g., regression coefficients), variances (e.g., group variances), and measures of association (e.g,. bivariate correlations). The statistical underpinnings are described in Mulder, Hoijtink, and Xin (2019) <arXiv:1904.00679>, Mulder and Gelissen (2019) <arXiv:1807.05819>, Mulder (2016) <DOI:10.1016/j.jmp.2014.09.004>, Mulder and Fox (2019) <DOI:10.1214/18-BA1115>, Mulder and Fox (2013) <DOI:10.1007/s11222-011-9295-3>, Boeing-Messing, van Assen, Hofman, Hoijtink, and Mulder <DOI:10.1037/met0000116>, Hoijtink, Mulder, van Lissa, and Gu, (2018) <DOI:10.31234/osf.io/v3shc>, Gu, Mulder, and Hoijtink, (2018) <DOI:10.1111/bmsp.12110>, Hoijtink, Gu, and Mulder, (2018) <DOI:10.1111/bmsp.12145>, and Hoijtink, Gu, Mulder, and Rosseel, (2018) <DOI:10.1037/met0000187>.

Regression for data too large to fit in memory. This package functions exactly like the 'biglm' package, but works with later versions of R.

Contains functions that lets you fit dynamic hazard models using state space models. The first implemented model is described in Fahrmeir (1992) <doi:10.1080/01621459.1992.10475232> and Fahrmeir (1994) <doi:10.1093/biomet/81.2.317>. Extensions hereof are available where the Extended Kalman filter is replaced by an unscented Kalman filter and other options including particle filters. The implemented particle filters support more general state space models.

Functions are provided for computing the density and the distribution function of d-dimensional normal and "t" random variables, possibly truncated (on one side or two sides, with componentwise choice), and for generating random vectors sampled from these distributions, except sampling from the truncated "t". Moments of arbitrary order of a truncated normal are computed, and converted to cumulants up to order 4. Probabilities are computed via non-Monte Carlo methods; different routines are used in the case d=1, d=2, d=3, d>3, if d denotes the dimensionality.

A collection of statistical tools for objective (non-supervised) applications of the Regional Frequency Analysis methods in hydrology. The package refers to the index-value method and, more precisely, helps the hydrologist to: (1) regionalize the index-value; (2) form homogeneous regions with similar growth curves; (3) fit distribution functions to the empirical regional growth curves. Most of the methods are those described in the Flood Estimation Handbook (Centre for Ecology & Hydrology, 1999, ISBN:9781906698003). Homogeneity tests from Hosking and Wallis (1993) <doi:10.1029/92WR01980> and Viglione et al. (2007) <doi:10.1029/2006WR005095> are available.