TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

The insignificance of statistical significance [pdf]

8 pointsby agconwayabout 14 years ago

1 comment

equarkabout 14 years ago
The problem with statistical significance is really not about weighing type I vs type II errors. It's much more about 1) basing decisions conditional on the data rather than the null, 3) dealing with nuisance parameters, 3) combining information from both from prior sources and through dependencies across unknown quantities and 4) having some flexible and coherent recipe for tackling inference when confronted with new applied problems.