# Dai Feng

#### 9 packages on CRAN

Bland-Altman plot and scatter plot with identity line for visualization and point and interval estimates for different metrics related to reproducibility/repeatability/agreement including the concordance correlation coefficient, intraclass correlation coefficient, within-subject coefficient of variation, smallest detectable difference, and mean normalized smallest detectable difference.

Estimate the AUC using a variety of methods as follows: (1) frequentist nonparametric methods based on the Mann-Whitney statistic or kernel methods. (2) frequentist parametric methods using the likelihood ratio test based on higher-order asymptotic results, the signed log-likelihood ratio test, the Wald test, or the approximate ''t'' solution to the Behrens-Fisher problem. (3) Bayesian parametric MCMC methods.

Various functions for random number generation, density estimation, classification, curve fitting, and spatial data analysis.

There are three sets of functions. The first produces basic properties of a graph and generates samples from multinomial distributions to facilitate the simulation functions (they maybe used for other purposes as well). The second provides various simulation functions for a Potts model in Potts, R. B. (1952) <doi:10.1017/S0305004100027419>. The third currently includes only one function which computes the normalizing constant of a Potts model based on simulation results.

Various algorithms for segmentation of 2D and 3D images, such as computed tomography and satellite remote sensing. This package implements Bayesian image analysis using the hidden Potts model with external field prior of Moores et al. (2015) <doi:10.1016/j.csda.2014.12.001>. Latent labels are sampled using chequerboard updating or Swendsen-Wang. Algorithms for the smoothing parameter include pseudolikelihood, path sampling, the exchange algorithm, approximate Bayesian computation (ABC-MCMC and ABC-SMC), and the parametric functional approximate Bayesian (PFAB) algorithm. Refer to <doi:10.1007/s11222-014-9525-6> and <doi:10.1214/18-BA1130> for further details.

Filtering, also known as gating, of flow cytometry samples using the curvHDR method, which is described in Naumann, U., Luta, G. and Wand, M.P. (2010) <DOI:10.1186/1471-2105-11-44>.

The Particle Gibbs sampler and Gibbs/Metropolis-Hastings sampler were implemented to fit Bayesian additive regression tree model. Construction of the model (training) and prediction for a new data set (testing) can be separated. Our reference papers are: Lakshminarayanan B, Roy D, Teh Y W. Particle Gibbs for Bayesian additive regression trees[C], Artificial Intelligence and Statistics. 2015: 553-561, <http://proceedings.mlr.press/v38/lakshminarayanan15.pdf> and Chipman, H., George, E., and McCulloch R. (2010) Bayesian Additive Regression Trees. The Annals of Applied Statistics, 4,1, 266-298, <doi:10.1214/09-aoas285>.