1Blurb::
2Uses Latin Hypercube Sampling (LHS) to sample variables
3
4Description::
5The \c lhs keyword invokes Latin Hypercube Sampling as the means of
6drawing samples of uncertain variables according to their probability
7distributions.  This is a stratified, space-filling approach that
8selects variable values from a set of equi-probable bins.
9
10<b> Default Behavior </b>
11
12Latin Hypercube Sampling is the default sampling mode in most contexts
13(exception: multilevel_sampling).  To explicitly specify LHS in the
14Dakota input file, the \c lhs keyword must appear in conjunction with
15the \c sample_type keyword.
16
17<b> Usage Tips </b>
18
19Latin Hypercube Sampling is very robust and can be applied to any
20problem.  It is fairly effective at estimating the mean of model
21responses and linear correlations with a reasonably small number of
22samples relative to the number of variables.
23
24Topics::
25Examples::
26\verbatim
27method
28  sampling
29    sample_type lhs
30    samples = 20
31\endverbatim
32Theory::
33Faq::
34See_Also::
35