README.md

Adaptive Rejection Sampling

R Package: Adaptive Reject Sampling (ars)

ars can quickly generates number of observations sampling from log-concave density function that is hard to evaluate using adaptive rejection sampling. For more details about the algorithms of adaptive rejection sampling, see Gilks et al. (1992). The implemented method is tangent approach.

Download the latest version

Development of ars can be tracked at https://github.com/yantingpan/ars. Package can be installed via devtools::install_github('yantingpan/ars')

Performance of sampling

Several log-concave density functions(trunacted normal, gamma, beta, etc.) are used to test the performance of ars. The density of resulting sample points genrated by ars fits well with the input density function.

Authors

Vincent Myers vincent_myers@berkeley.edu, Yanting Pan yanting_pan@berkeley.edu, Zhenni Ye ye.zhenni@berkeley.edu. [For any mistakes found, please accept our apology and email Yanting through yanting_pan@berkeley.edu so we can fix it as soon as possible.]



yantingpan/ars documentation built on May 21, 2019, 10:14 a.m.