GpSelection.jl is a Julia package for Gaussian process selections in semiparametric regression for multi-pathway analysis.
To download the package, use the code below.
Pkg.clone("git@github.com:jlin-vt/GpSelection.jl.git")
Pkg.build("GpSelection")
n = 100; K = 5; num_sig = 2; df = 1.0; p_beta = 2
dat = generateData(K, p_beta, n, df, num_sig)
options = Options(100)
priors = PriorPars(5, 10, 10, fill(5, K), fill(10, K), 0.0, zeros(K, K))
@elapsed beta_E, sigma2_E, gamma_E, H_E = varBayes(dat, priors, options)
The preprint describing the corncob methodology is available here. The manuscript has been submitted to Journal of Statistical Computation and Simulation.
If you encounter a bug or would like make a change request, please file it as an issue here.
The GpSelection.jl package is licensed under the MIT "Expat" License.