A variable transformation of a low-discrepancy uniform design yields low discrepancy for a target distribution only if the discrepancy kernels satisfy certain conditions; remedies include optimal one-dimensional projections for dense designs and coordinate-exchange optimization for any design size.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.CO 1years
2020 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Is a Transformed Low Discrepancy Design Also Low Discrepancy?
A variable transformation of a low-discrepancy uniform design yields low discrepancy for a target distribution only if the discrepancy kernels satisfy certain conditions; remedies include optimal one-dimensional projections for dense designs and coordinate-exchange optimization for any design size.