random_start: Generate a random initial starting point for EM algorithm

In emery: Accuracy Statistic Estimation for Imperfect Gold Standards

Generate a random initial starting point for EM algorithm

Description

random_start() is a general function for creating a random, plausible starting point for the EM algorithm to begin iterating from. Varying the starting points of repeated calculations can help the researcher detect local extremes.

Creates random initial sensitivity and specificity values for n_method methods. Values are generated from a random beta distribution with shape parameters a=3 and b=1. Prevalence is a random value from a uniform distribution.

Usage

random_start(type, n_method, method_names)

random_start_binary(n_method = NULL, method_names = NULL)

random_start_ordinal(n_method = NULL, method_names = NULL)

Arguments

type	A string specifying the data type of the methods being simulated.
n_method	An integer representing the number of methods to simulate.
method_names	Optional vector of names used to identify each method.

Value

List containing initial values to be passed to init argument

emery documentation built on Feb. 23, 2026, 1:06 a.m.