Dakota Reference Manual  Version 6.4
Large-Scale Engineering Optimization and Uncertainty Analysis
 All Pages
logit_transform


Utilize the logit transformation to reduce sample rejection for bounded domains

Specification

Alias: none

Argument(s): none

Description

The logit transformation performs an internal variable transformation from bounded domains to unbounded domains in order to reduce sample rejection due to an out-of-bounds condition.

Default Behavior

This option is experimental at present, and is therefore defaulted off.

Usage Tips

This option can be helpful when regions of high posterior density exist in the corners of a multi-dimensional bounded domain. In these cases, it may be difficult to generate feasible samples from the proposal density, such that transformation to unbounded domains may greatly reduce sample rejection rates.

Examples

method,
        bayes_calibration queso
          samples = 2000 seed = 348
          dram
          logit_transform