Timeline for Cumming (2008) claims that distribution of p-values obtained in replications depends only on the original p-value. How can it be true?
Current License: CC BY-SA 3.0
13 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Dec 19, 2016 at 22:27 | comment | added | amoeba | Update: I read Goodman's and Senn's papers linked above and have now posted my own answer to summarize my current intuition. (By the way, I was happy to accept your answer and award it the bounty. Thanks again.) | |
| Dec 19, 2016 at 21:06 | vote | accept | amoeba | ||
| Dec 17, 2016 at 1:08 | history | bounty ended | amoeba | ||
| Dec 16, 2016 at 22:32 | history | edited | GeoMatt22 | CC BY-SA 3.0 |
add note on meta-analysis
|
| Dec 14, 2016 at 20:29 | comment | added | GeoMatt22 | amoeba thank you for digging up these references, they look interesting! For completeness, I added a "discussion" section trying to connect the Cumming result and the standard framework. | |
| Dec 14, 2016 at 20:28 | history | edited | GeoMatt22 | CC BY-SA 3.0 |
add discussion on Gaussian prior to connect end-members
|
| Dec 14, 2016 at 20:23 | history | edited | GeoMatt22 | CC BY-SA 3.0 |
add discussion on Gaussian prior to connect end-members
|
| Dec 14, 2016 at 18:04 | comment | added | amoeba | I found some old and not so old discussions of this topic: Goodman 1992, a comment on Goodman by Senn 2002, and a recent Lazzeroni et al 2014. The last one seems rather unhelpful (but I mention it for completeness) but the first two, particularly Senn's comment, appear very pertinent. | |
| Dec 14, 2016 at 17:23 | history | edited | GeoMatt22 | CC BY-SA 3.0 |
add note on Jeffreys prior
|
| Dec 14, 2016 at 16:58 | history | edited | GeoMatt22 | CC BY-SA 3.0 |
add summary
|
| Dec 14, 2016 at 16:49 | comment | added | GeoMatt22 | My hope is that by exposing the underlying assumption (e.g. uniform prior on hidden parameter), the discussion can now focus on the scientific/statistical question that I believe was your target! (Rather than the math/probability question I answered above.) | |
| Dec 14, 2016 at 15:55 | history | edited | GeoMatt22 | CC BY-SA 3.0 |
add comparison plot
|
| Dec 14, 2016 at 6:08 | history | answered | GeoMatt22 | CC BY-SA 3.0 |