Learn R Programming

jaatha (version 2.7.0)

Jaatha.initialSearch: Search the parameter space for good starting positions

Description

This functions divides the parameter space in different parts (blocks). In each block, simulations for different parameter combinations are run to roughly predict the combination with the highest score (which is equivalent to the highest composite log likelihood). This points can later be used as starting positions for the second estimation phase of Jaatha (Jaatha.refinedSearch)

Usage

Jaatha.initialSearch(jaatha, sim = 200, blocks.per.par = 2, rerun = FALSE)

Arguments

jaatha
The Jaatha settings (create with Jaatha.initialize)
sim
An integer stating the number of simulations that are performed for each block.
blocks.per.par
Integer. For each parameter axis in turn, we divide the parameter space into blocks.per.par equally sized blocks by restricting them to only a fraction of this axis.
rerun
You can repeat a previously done initial search in Jaatha. Do do so, just call the initial search function with the jaatha object result of the first initial search and set rerun to 'TRUE'.

Value

  • The jaatha object with starting positions