#pvalue

Dr Mircea Zloteanu ☀️ 🌊🌴mzloteanu
2025-07-02

#378 Selective Inference: The Silent Killer of Replicability

Thoughts: Benjamin overviews the replicability crisis, alternatives to p-values (and their issues), and suggests selective reporting is a large issue itself.

hdsr.mitpress.mit.edu/pub/l39r

Dr Mircea Zloteanu ☀️ 🌊🌴mzloteanu
2025-06-12
Dr Mircea Zloteanu ☀️ 🌊🌴mzloteanu
2025-06-11

#363 p-checker The one-for-all p-value analyzer

Thoughts: Easy way to check for publication bias using some current tools.

shinyapps.org/apps/p-checker/

Dr Mircea Zloteanu ☀️ 🌊🌴mzloteanu
2025-06-10
Dr Mircea Zloteanu ☀️ 🌊🌴mzloteanu
2025-06-09

#361 Vovk-Sellke Maximum p-Ratio (VS-MPR)

Thoughts: Can p-values provide evidence for H1 over H0? YES!*
(*if you make some assumptions and go a smidgen Bayesian)

VS-MPR is reported in @JASPStats

shinyapps.org/apps/vs-mpr/

Dr Mircea Zloteanu ☀️ 🌊🌴mzloteanu
2025-05-19

#346 Jeffreys-Lindley paradox

Thoughts: I like this short explanation of the "paradox" of why frequentist and bayesian inference can differ.

michael-franke.github.io/intro

2025-04-30

Пять ошибок при A/B-тестировании

Привет, Хабр! В A/B тестах можно ошибиться ещё до того, как первая строчка кода теста будет написана. А последствия этих ошибок сказываются не только на результатах одного эксперимента, а на всей продуктовой стратегии компании: на найме, на развитии функциональности, на распределении инвестиций. Сегодня рассмотрим пять системных ошибках, которые делают ваши A/B тесты ненадёжными — даже если снаружи всё выглядит корректно.

habr.com/ru/companies/otus/art

#python #AB_тесты #статистическая_значимость #выбор_метрики #когортный_анализ #pvalue

2025-04-17

For anyone still submitting biology papers with p-values:

"Estimation statistics should replace significance testing", Claridge-Chang and Assam, 2016
nature.com/articles/nmeth.3729 (and also at zenodo.org/records/60156 ).

#statistics #SignificanceTesting #pvalue #ScientificPublishing

Bjørn Sætreviksatrevik@fediscience.org
2025-03-25

Imagine you preregistered study with tests A and B for a directed hypothesis H, but did not specify one-tailed or two-tailed testing. Test A is in the predicted direction, but p = between .05 and .09. Test B is in the non-predicted direction, but p < .05. How do you report results?

#PreRegistration #OpenScience #NHST #SignificanceTesting #pValue

elaprendiz0000elaprendiz0000
2024-12-28

Explora los conceptos de Chi-Cuadrado y P-Value con un ejemplo práctico. Aprende a evaluar asociaciones entre variables categóricas. ¡No te lo pierdas! 👇

🔗 blogger.com/blog/post/edit/723

Jan R. Boehnkejrboehnke
2024-08-28

Best moment in the video:
"I'm not gonna define p-values today, forgive me."

Maybe James should.
Not that there aren't any good definitions (e.g. check ncbi.nlm.nih.gov/pmc/articles/).

But presented by James, at least coffee drinkers might understand it (better)! 😅

2024-06-25

Об ускорении некоторых тестов на нормальность из библиотеки SciPy

Приветствую уважаемых читателей, случайно или нет наткнувшихся на эту статью. Адресована она всем тем, кто исследует распределение разнообразных эмпирических данных. В моей публикации вас ждут: небольшой ликбез по теории вероятностей, развивающий её интуитивное и практическое понимание; детективная история о том, как решение дифференциального уравнения привело нас к двум важнейшим статистическим характеристикам выборочных последовательностей; ревизионизм в области проверки распределений на соответствие гауссовскому и объяснение пресловутого p-value с нуля за 5 минуты.

habr.com/ru/companies/aktiv-co

#scipy #пирсон #pvalue #нормальное_распределение #эксцесс

2023-10-04

It's been one of those days in Statsland, folks

#statistics #pvalue

Inigo Montoya meme reading "p-value, I do not think it means what you think it means"
2023-09-12

I used to be annoyed by the absurdly low p-values that can be obtained when doing #correlations with thousands of data points. Permutation analysis could help to see if the observed relationships might be a random occurence.

Searching now for tools that automate the process, while also explaining clearly what is being done.

#statistics #pvalue #research

Nick Byrd, Ph.D.ByrdNick@nerdculture.de
2023-06-02

"A Sensitivity Test Does Everything That a Significance Test Does, And Better". The former
- uses the same information as the latter's #pValue
- doesn't assume perfect randomization or measurement
- can deal with missing data

...and unless #stats software offers sensitivity analysis by default, the status quo will continue.

PDF: iosrjournals.org/iosr-jrme/pap

Issue: iosrjournals.org/iosr-jrme/pag

Dr Mircea Zloteanu ☀️ 🌊🌴mzloteanu
2023-03-02

New blogpost 🚨 : What do you do if you found a significant result, but your study was underpowered? How reliable is your finding? I discuss Type M and Type S error

mzstats.blogspot.com/2023/02/w

2023-02-02

Came across this interesting paper on the limitations of permutation as a method for computing the p-value of a test: academic.oup.com/bioinformatic
#stats #pvalue #permutation

Dr Mircea Zloteanu ☀️ 🌊🌴mzloteanu
2023-01-14

Part II of my series on what to do with non-significant results is up now.

In this post, I focus on how to determine if your data is compatible with the claim of "no effect" (and why relying on p-values is wrong). It covers TOST equivalence tests and how to interpret and write up your results for publication. Link below 🔗

mzstats.blogspot.com/2023/01/w

Dr Mircea Zloteanu ☀️ 🌊🌴mzloteanu
2023-01-08

Everyone focuses on misunderstandings of p-values, but the bigger problem is understanding effect sizes and power!

If you ran a properly powered study and were aware of a reasonable effect size in your research, then misinterpreting the p-value in your manuscript at the end will affect the scientific field less, as others can correct the interpretation later.

Client Info

Server: https://mastodon.social
Version: 2025.04
Repository: https://github.com/cyevgeniy/lmst