Thresholding the null-conditional probability that a fixed-sample test would reject at its planned endpoint gives an anytime-valid (up to the maximum) sequential test with controlled Type I error.
Title resolution pending
1 Pith paper cite this work, alongside 665 external citations. Polarity classification is still indexing.
1
Pith paper citing it
665
external citations · OpenAlex
fields
stat.ME 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Predicting fixed-sample test decisions enables anytime-valid inference
Thresholding the null-conditional probability that a fixed-sample test would reject at its planned endpoint gives an anytime-valid (up to the maximum) sequential test with controlled Type I error.