optimize_pi

This is an internal (non-exported) function. This help
 page provides additional documentation mainly intended for
 developers and expert users.

internal

Implements the multivariate adaptive shrinkage (mash)
method of Urbut et al (2019) <DOI:10.1038/s41588-018-0268-8> for
estimating and testing large numbers of effects in many conditions
(or many outcomes). Mash takes an empirical Bayes approach to
testing and effect estimation; it estimates patterns of similarity
among conditions, then exploits these patterns to improve accuracy
of the effect estimates. The core linear algebra is implemented in
C++ for fast model fitting and posterior computation.

Peter Carbonetto

mashr

Multivariate Adaptive Shrinkage

Matthew Stephens

Sarah Urbut

Gao Wang

Yuxin Zou

Yunqi Yang

Sam Roweis

David Hogg

Jo Bovy

optimize_pi function

<dl><dt>matrix_lik</dt>
<dd>a matrix of likelihoods, where the (i,k)th entry
is the probability of observation i given it came from component k
of g</dd>
<dt>pi_init</dt>
<dd>numeric vector specifying value from which to
initialize optimization</dd>
<dt>prior</dt>
<dd>numeric vector specifying prior to use in the
penalized likelihood</dd>
<dt>optmethod</dt>
<dd>a string, giving name of optimization function to
use</dd>
<dt>control</dt>
<dd>a list of parameters to be passed to optmethod</dd></dl>

Arguments

Estimate mixture weights by maximum (penalized) likelihood — optimize_pi

<dl>

<dt>matrix_lik</dt>
<dd>a matrix of likelihoods, where the (i,k)th entry
is the probability of observation i given it came from component k
of g</dd>


<dt>pi_init</dt>
<dd>numeric vector specifying value from which to
initialize optimization</dd>


<dt>prior</dt>
<dd>numeric vector specifying prior to use in the
penalized likelihood</dd>


<dt>optmethod</dt>
<dd>a string, giving name of optimization function to
use</dd>


<dt>control</dt>
<dd>a list of parameters to be passed to optmethod</dd>

</dl>

Estimate mixture weights by maximum (penalized) likelihood

optimize_pi: Estimate mixture weights by maximum (penalized) likelihood

Description

Usage

Value

Arguments