The probability distribution of p values
This is a companion to the short paper at
https://arxiv.org/pdf/1603.07532.pdf
It explains the problem with p-values giving different values across statistically identical samples.
Sorry for the lighting but I am not into these things.
Abstract: We present an explicit and parsimonious probability distribution (meta-distribution) for p-values across ensembles of statistically identical phenomena, having for sole parameter the median "true" p-value. P-values are extremely skewed and volatile, regardless of the sample size n, and vary greatly across repetitions of exactly same protocols under identical stochastic copies of the phenomenon. The convenience of formula allows the investigation of scientific results, particularly meta-analyses, and the expression of such biases as "p-hacking".
---
Typo: median p is not "true p" (skewness of p) but a proxy.
Видео The probability distribution of p values канала N N Taleb's Probability Moocs
https://arxiv.org/pdf/1603.07532.pdf
It explains the problem with p-values giving different values across statistically identical samples.
Sorry for the lighting but I am not into these things.
Abstract: We present an explicit and parsimonious probability distribution (meta-distribution) for p-values across ensembles of statistically identical phenomena, having for sole parameter the median "true" p-value. P-values are extremely skewed and volatile, regardless of the sample size n, and vary greatly across repetitions of exactly same protocols under identical stochastic copies of the phenomenon. The convenience of formula allows the investigation of scientific results, particularly meta-analyses, and the expression of such biases as "p-hacking".
---
Typo: median p is not "true p" (skewness of p) but a proxy.
Видео The probability distribution of p values канала N N Taleb's Probability Moocs
Показать
Комментарии отсутствуют
Информация о видео
23 марта 2016 г. 7:54:07
00:09:41
Другие видео канала
![MINI LECTURE 17: Maximum Ignorance Probability (a bit more technical)](https://i.ytimg.com/vi/lnaxM4nWWBM/default.jpg)
![MINI LECTURE 15 - Conditional vs. unconditional correlation: twin studies overestimate heredity.](https://i.ytimg.com/vi/azdJAHF2Mgw/default.jpg)
![MINI-LESSON 4: CLT, The Central Limit Theorem, a nontechnical presentation.](https://i.ytimg.com/vi/bfM9efdStN8/default.jpg)
![MINI LESSON 11: Path dependence in 2 ½ minutes](https://i.ytimg.com/vi/zLs6ggd3BPo/default.jpg)
![How to bluff when you don't know what you are talking about.](https://i.ytimg.com/vi/RSSYkb_dmt8/default.jpg)
![MINI-LESSON 2: Fat Tails, a Very, Very Introductory Presentation.](https://i.ytimg.com/vi/t7Fr6iGhmBM/default.jpg)
![Stephen Wolfram visits RWRI 18 (Summer Workshop)](https://i.ytimg.com/vi/60sNXOyGmps/default.jpg)
![Weak and Strong Non-Ergodicities](https://i.ytimg.com/vi/qA_6BWkC4og/default.jpg)
![MINI LESSON 11b: Drawdowns and Logs](https://i.ytimg.com/vi/w-LVLIwkqlk/default.jpg)
![How Increasing Benefits Increases the Risk of Ruin](https://i.ytimg.com/vi/FoP32AesRzE/default.jpg)
![MINI-LESSON 6: Fooled by Metrics (Correlation)](https://i.ytimg.com/vi/fb921ZrM6h0/default.jpg)
![MINI-LESSON 8: Power Laws (maximally simplified)](https://i.ytimg.com/vi/oMl-SbuQUYc/default.jpg)
![Jouer sa peau](https://i.ytimg.com/vi/pzBdTV8Qwo8/default.jpg)
![QUANTITATIVE FINANCE 2: We don't use Black-Scholes, the simpler derivation vindicating Bachelier.](https://i.ytimg.com/vi/Au_83c49dHA/default.jpg)
![MINI LECTURE 18: How to build a positive definite correlation matrix for Monte Carlo simulations](https://i.ytimg.com/vi/smtSSYjJ8mo/default.jpg)
![Naive empiricism, terrorism & Ebola.](https://i.ytimg.com/vi/9dKiLclupUM/default.jpg)
![QUANTITATIVE FINANCE 1: Deriving Black-Sholes via Itô's lemma (the dynamic hedging approach)](https://i.ytimg.com/vi/YizR8uLlCes/default.jpg)
![MINI-LESSON 3: The Law of Large Numbers. A very intuitive introduction.](https://i.ytimg.com/vi/zyBXNIskQK0/default.jpg)
![The Randomness of Correlation and its Hacking by Bigdataists](https://i.ytimg.com/vi/yUooqL47akM/default.jpg)
![From the Convexity of Masks to No-Convexity Ben](https://i.ytimg.com/vi/L5DCWU6heKI/default.jpg)