aCSS-B performs approximately valid frequentist goodness-of-fit tests by conditioning on posterior samples as an approximate sufficient statistic, expanding the method to three common null models where existing techniques fail.
After the burn-in ofB 0 = 500, we extractB= 25 posterior samples{t (b), γ(b)}at every tenth step
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.ME 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Conditioning on posterior samples for flexible frequentist goodness-of-fit testing
aCSS-B performs approximately valid frequentist goodness-of-fit tests by conditioning on posterior samples as an approximate sufficient statistic, expanding the method to three common null models where existing techniques fail.