find_optimum: Routine for finding MASSIVE posterior local optimum.

In igbucur/MASSIVE: MASSIVE: Tractable and Robust Bayesian Learning of Many-Dimensional Instrumental Variable Models

Description

Routine for finding MASSIVE posterior local optimum.

Usage

find_optimum(
  J,
  N,
  SS,
  sigma_G,
  prior_sd,
  skappa_X,
  skappa_Y,
  post_fun = scaled_neg_log_posterior,
  gr_fun = scaled_neg_log_gradient,
  hess_fun = scaled_neg_log_hessian
)

Arguments

J	Integer number of genetic instrumental variables.
N	Integer number of observations.
SS	Numeric matrix containing first- and second-order statistics.
sigma_G	Numeric vector of genetic IV standard deviations.
prior_sd	List of standard deviations for the parameter Gaussian priors.
skappa_X	Scale-free confounding coefficient to exposure used for initialization.
skappa_Y	Scale-free confounding coefficient to outcome used for initialization.
post_fun	Function for computing the IV model posterior value.
gr_fun	Function for computing the IV model posterior gradient.
hess_fun	Function for computing the IV model posterior Hessian.

Value

optim object (see optim containing optimum parameters, the value obtained at the optimum, and the Hessian at the optimum.

Examples

J <- 5 # number of instruments
N <- 1000 # number of samples
parameters <- random_Gaussian_parameters(J) 
EAF <- runif(J, 0.1, 0.9) # EAF random values
dat <- generate_data_MASSIVE_model(N, 2, EAF, parameters)
find_optimum(
  J, N, dat$SS, binomial_sigma_G(dat$SS), 
  prior_sd = decode_IV_model(get_random_IV_model(J), 1, 0.01), 
  skappa_X = 1, skappa_Y = 1
)

igbucur/MASSIVE documentation built on Oct. 26, 2020, 1:26 a.m.