adaLASSODC: Adaptive Bayesian Lasso with Unpenalized Design Covariates

In abnormally-distributed/Bayezilla: Bayesian Models With JAGS and Several Utilities for Data Exploration and Model Evaluation.

Description

The Adaptive Bayesian LASSO of Leng, Tran and David Nott (2018).This function has the further allowance for a set of covariates that are not penalized. For example, you may wish to include variables such as age and gender so that the coefficients for the other variables are penalized while controlling for these. This is a common need in research. Basically just the Bayesian LASSO of Park & Casella (2008) but with individual lambdas on each parameter defined by a gamma(sh, ra) distribution, where sh and ra are shape and rate hyperparameters. Here sh and ra are given independent gamma(4, 8) and gamma(1, 5) priors respectively. This places the expected values for the shape and rate parameters at 0.50 and 0.20 respectively, which is consistent with the gamma(0.25, 0.20) prior on lambda used for most other shrinkage models with additional design covariates in this package. For the binomial and poisson likelihood functions the uniform-gamma scale mixture for the variant of the Bayesian LASSO is adapted for use here.

For alternatives see negLASSO (which is extremely similar) or extLASSO.

Model Specification:



Plugin Pseudo-Variances:

Usage

adaLASSODC(formula, design.formula, data, family = "gaussian",
  log_lik = FALSE, iter = 10000, warmup = 1000, adapt = 2000,
  chains = 4, thin = 1, method = "parallel", cl = makeCluster(2),
  ...)

Arguments

formula	the model formula
design.formula	formula for the design covariates.
data	a data frame.
family	one of "gaussian", "binomial", or "poisson".
log_lik	Should the log likelihood be monitored? The default is FALSE.
iter	How many post-warmup samples? Defaults to 10000.
warmup	How many warmup samples? Defaults to 1000.
adapt	How many adaptation steps? Defaults to 2000.
chains	How many chains? Defaults to 4.
thin	Thinning interval. Defaults to 1.
method	Defaults to "parallel". For an alternative parallel option, choose "rjparallel" or. Otherwise, "rjags" (single core run).
cl	Use parallel::makeCluster(# clusters) to specify clusters for the parallel methods. Defaults to two cores.
...	Other arguments to run.jags.

Value

a runjags object

References

Park, T., & Casella, G. (2008). The Bayesian Lasso. Journal of the American Statistical Association, 103(482), 681-686. Retrieved from http://www.jstor.org/stable/27640090

Mallick, H., & Yi, N. (2014). A New Bayesian Lasso. Statistics and its interface, 7(4), 571–582. doi:10.4310/SII.2014.v7.n4.a12

Leng, C., Tran, M.N., & Nott, D.J. (2014). Bayesian adaptive Lasso. arXiv:1009.2300

Examples

adaLASSODC()

abnormally-distributed/Bayezilla documentation built on Oct. 31, 2019, 1:57 a.m.