Timeline for What does Bayesian Hypothesis Testing mean in the framework of inference and decision theory?
Current License: CC BY-SA 3.0
22 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Apr 13, 2017 at 12:44 | history | edited | CommunityBot |
replaced http://stats.stackexchange.com/ with https://stats.stackexchange.com/
|
|
| S Jan 4, 2015 at 2:15 | history | bounty ended | Charlie Parker | ||
| S Jan 4, 2015 at 2:15 | history | notice removed | Charlie Parker | ||
| Jan 2, 2015 at 22:53 | history | edited | Charlie Parker | CC BY-SA 3.0 |
added 711 characters in body
|
| Dec 28, 2014 at 22:37 | answer | added | M T | timeline score: 3 | |
| S Dec 28, 2014 at 21:36 | history | suggested | M T | CC BY-SA 3.0 |
Fixed some minor grammar but mainly wanted to make clear the distinction between Khan's frequentist example and the Bayesian approach. It seemed a couple well chosen additions to the question would be much simpler than having to quote it. I will add a more detailed comment.
|
| Dec 28, 2014 at 21:19 | review | Suggested edits | |||
| S Dec 28, 2014 at 21:36 | |||||
| Dec 28, 2014 at 14:59 | answer | added | Livid | timeline score: 0 | |
| Dec 28, 2014 at 10:04 | comment | added | Xi'an | The first line of my Answer is "A statistical model is given by a family of probability distributions". The pair (distribution,statistical model) does not make sense. And yes indeed the Bayesian approach puts a prior on the pair (model index, model parameter), see line 9 of my Answer. | |
| Dec 28, 2014 at 9:57 | comment | added | Xi'an | My answer uses the normal example, which is one of the simplests I can think of: $H_0:\,X\sim\mathcal{N}(0,1)$ versus $H_1:\,X\sim\mathcal{N}(\theta,1)$. If this setting does not make sense to you, I strongly suggest you read an introductory textbook (as for instance this free on-line version). | |
| Dec 28, 2014 at 5:59 | comment | added | Charlie Parker | @rocinante is there a reference where I can see your coin hypothesis example explained in both paradigms? Is it to much to ask for you to explain it as an answer (if you can/want, I would appreciate it (and probably reward it), but I understand it can be annoying)? Thanks for your time and book suggestion, I am excited to read it! :) | |
| Dec 28, 2014 at 4:17 | comment | added | rocinante | 2/2 Suppose you have a coin and you want to see if it is fair, so you flip it 50 times. You now have a data set about which you want to make some inference (i.e. is the coin biased or not). Logically, if the coin is fair, about half the tosses should be heads. (Note that this is not a stats derivation, but your own logical reasoning). That is your hypothesis. You can test this hypothesis 2 ways: the Bayesian way and the frequentist way. | |
| Dec 28, 2014 at 4:14 | history | tweeted | twitter.com/#!/StackStats/status/549055819955048448 | ||
| Dec 28, 2014 at 4:14 | comment | added | rocinante | It's not an easy thing to understand because it's not an easy thing to articulate in a concise way. Rather than think about this in abstract terms (like maps), maybe it will help if you think about it with a simpler example.1/2 | |
| Dec 28, 2014 at 4:01 | comment | added | Charlie Parker | @rocinante I think I agree with you. I am definitively confused about hypothesis testing in general (and the bayesian framework doesn't help at all). I will definitively take a look at that. Thanks for your patience and understanding, its greatly appreciated. | |
| Dec 28, 2014 at 3:04 | comment | added | rocinante | I am hesitant to wade into this discussion because I think your problem is really that one of understanding what hypothesis testing means in principle, rather than specifically what hypothesis testing is in the Bayesian framework. To help with this, I suggest having a look at the book "Modes of Parametric Statistical Inference" by Geisser. books.google.ca/… | |
| S Dec 28, 2014 at 2:31 | history | bounty started | Charlie Parker | ||
| S Dec 28, 2014 at 2:31 | history | notice added | Charlie Parker | Improve details | |
| Dec 23, 2014 at 18:33 | comment | added | Charlie Parker | @Xi'an I read the following wikipedia article: en.wikipedia.org/wiki/Statistical_model is that what they mean by a model and a hypothesis? thnx for ur patience btw :) | |
| Dec 23, 2014 at 18:30 | answer | added | Xi'an | timeline score: 10 | |
| Dec 23, 2014 at 5:08 | history | edited | Charlie Parker | CC BY-SA 3.0 |
added 145 characters in body
|
| Dec 23, 2014 at 2:00 | history | asked | Charlie Parker | CC BY-SA 3.0 |