gmo: Gradient-based marginal optimization
In stan-dev/gmo: Gradient-based Marginal Optimization

Description Usage Arguments Value

Maximize the marginal posterior with respect to specified parameters, with nuisance parameters marginalized out.

1
2
3

gmo(file, local_file, full_model, data, method = c("laplace"),
  init = "random", draws = 5L, iter = 100L, inner_iter = 10L,
  cond_iter = NA, eta = 1, tol = 0.001, seed = 1234L)

`file`	A character string or a connection that R supports specifying the Stan model specification in Stan's modeling language.
`local_file`	A character string or a connection that R supports specifying the Stan model specification in Stan's modeling language.
`full_model`	If provided, an object of class 'stanfit' that makes it unnecessary to pass 'file' or 'local_file'
`data`	A named ‘list’ or ‘environment’ providing the data for the model or a character vector for all the names of objects used as data. See `stan` for more details.
`method`	A character string naming the conditional inference: "laplace"
`init`	A numeric vector of length the number of hyperparameters.
`draws`	A positive integer, number of draws to calculate stochastic gradient.
`iter`	A positive integer, the maximum number of iterations.
`inner_iter`	A positive integer, the number of iterations after each conditional inference.
`cond_iter`	A positive integer, the maximum number of iterations for the conditional inference. Default is to run until convergence.
`eta`	Double, constant scale factor for learning rate.
`tol`	Double, tolerance for signaling convergence.
`seed`	The seed, a positive integer, for random number generation of Stan. The default is generated from 1 to the maximum integer supported by R so fixing the seed of R's random number generator can essentially fix the seed of Stan. When multiple chains are used, only one seed is needed, with other chains' seeds being generated from the first chain's seed to prevent dependency among the random number streams for the chains. When a seed is specified by a number, ‘as.integer’ will be applied to it. If ‘as.integer’ produces ‘NA’, the seed is generated randomly. We can also specify a seed using a character string of digits, such as ‘"12345"’, which is converted to integer.

An object of reference class "gmo". It is a list containing the following components:

`par`	a vector of optimized parameters
`cov`	estimated covariance matrix at `par`
`sims`	`draws * inner_iter` many samples from the last approximation to the conditional posterior, p(alpha \| y, phi)