tag:blogger.com,1999:blog-16612221.post3501331813041589137..comments2024-03-19T03:14:04.172-04:00Comments on Rabett Run: On Priors, Bayesians and FrequentistsEliRabetthttp://www.blogger.com/profile/07957002964638398767noreply@blogger.comBlogger14125tag:blogger.com,1999:blog-16612221.post-66284145368333834742013-02-09T01:48:46.261-05:002013-02-09T01:48:46.261-05:00'frequantism is a figment of a quantum mind.&#...'frequantism is a figment of a quantum mind.', anonymous #743.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-16612221.post-35588339461710823382013-02-08T22:43:20.729-05:002013-02-08T22:43:20.729-05:00Socrates long history as a proponent of burning gl...Socrates long history as a proponent of burning glass SRM does not inspire confidence-- <br /><br />There is rumor in Oxyrhinchus of a lost play of Aristophane, s in which the philosopher refuses to whitewash his house because he thinks the Athenian Heat Island Effect has spoiled weather reports from the Tower of the Winds.THE CLIMATE WARShttps://www.blogger.com/profile/02578106673226403151noreply@blogger.comtag:blogger.com,1999:blog-16612221.post-30297821015874961332013-02-08T21:31:12.391-05:002013-02-08T21:31:12.391-05:00I read
Deborah G. Mayo
Statistical Science and Ph...I read<br /><br />Deborah G. Mayo<br />Statistical Science and Philosophy of Science: Where Do/Should They Meet in 2011 (and Beyond)?<br />RMM Vol. 2, 2011, 79–102<br /><br />and<br /><br />Deborah G. Mayo<br />Statistical Science and Philosophy of Science Part 2: Shallow versus Deep Explorations<br />RMM Vol. 3, 2012, 71–107.<br /><br />Not impressed.<br /><br />Also read<br /><br />Bradley Efron<br />A 250-year argument: Belief, behavior and the bootstrap<br />Bull. AMS 50:1, Jan 2013, 129--146.<br /><br />Impressed.David B. Bensonhttps://www.blogger.com/profile/02917182411282836875noreply@blogger.comtag:blogger.com,1999:blog-16612221.post-27381751026677934862013-02-08T19:20:13.288-05:002013-02-08T19:20:13.288-05:00The whole issue seems to be Open Access:
http://o...The whole issue seems to be Open Access:<br /><br />http://onlinelibrary.wiley.com/doi/10.1111/bmsp.2013.66.issue-1/issuetoc<br /><br />So we can read Mayo's criticisms and the authors' rejoinder.willardhttp://neverendingaudit.tumblr.comnoreply@blogger.comtag:blogger.com,1999:blog-16612221.post-22700964690261617882013-02-08T19:00:10.075-05:002013-02-08T19:00:10.075-05:00Bill, would you prefer, ////Eli=Bandit, Indiana=Wm...Bill, would you prefer, ////Eli=Bandit, Indiana=Wm Briggs or Steve JewsonAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-16612221.post-37328147004173422992013-02-08T18:54:57.544-05:002013-02-08T18:54:57.544-05:00Andrew Gelman reports reviews for his article with...Andrew Gelman reports reviews for his article with Shalizi, handwaved in the dialog:<br /><br />http://andrewgelman.com/2013/02/philosophy-and-the-practice-of-bayesian-statistics-with-discussion/<br /><br />More backreading for Socrates!willardhttp://neverendingaudit.tumblr.comnoreply@blogger.comtag:blogger.com,1999:blog-16612221.post-83017963569834690182013-02-07T21:26:21.069-05:002013-02-07T21:26:21.069-05:00My my, this anonymouse is all by way of mouth and ...My my, this anonymouse is all by way of mouth and none by way of trousers...<br /><br />PS An <i>Indian village</i> in RoLA? You can't even get that right!billnoreply@blogger.comtag:blogger.com,1999:blog-16612221.post-67459779442330239422013-02-07T20:48:19.565-05:002013-02-07T20:48:19.565-05:00Eli, you say, " Real science is messy, this i...Eli, you say, " Real science is messy, this is arguing for only doing things when you know the answer before you start. Is statistics a tool or a means in itself, if it is a tool, why let it run your life?" <br /><br />Might be messy Eli, but you still have to do it - and you ain't doing it. This post reminds me of Raiders of the Lost Ark, where in the center of an Indian village a knife wielding bandit confronts Indiana in a highly contorted visual display. Indiana pulls out his Colt and promptly disposes of the bandit. Bandit=Eli, Indiana=VS.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-16612221.post-27056260367248570842013-02-07T11:26:08.904-05:002013-02-07T11:26:08.904-05:00Kruger, and note several more recent papers (with ...Kruger, and note several more recent papers (with various authors) extend the findings of the eponymous paper.<br /><br />E.g.<br />http://www.sciencedirect.com/science/article/pii/S074959780700060XHank Robertshttps://www.blogger.com/profile/07521410755553979665noreply@blogger.comtag:blogger.com,1999:blog-16612221.post-13521329243076795132013-02-07T09:56:06.782-05:002013-02-07T09:56:06.782-05:00Michael Jordan might appreciate if you include cal...Michael Jordan might appreciate if you include calibration and coherence in your limerick, Horatio.<br /><br />See his slide after his explanation in decision-theoretic terms.willardhttp://neverendingaudit.tumblr.comnoreply@blogger.comtag:blogger.com,1999:blog-16612221.post-2283558681516600922013-02-07T08:08:32.952-05:002013-02-07T08:08:32.952-05:00"Probably Wrong"
-- by Horatio Algeranon..."Probably Wrong"<br />-- by Horatio Algeranon<br /><br />Bayesians are frequently wrong<br />When priors are unreasonably strong.<br />Frequentists are oft unfazed <br />When models are wrongly bayesed.<br />Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-16612221.post-88412597550824269032013-02-07T04:56:29.852-05:002013-02-07T04:56:29.852-05:00If you are consistent, your prior will reflect you...If you are consistent, your prior will reflect your past evidence. Strong contradiction, in the sense of having strong evidence for A first, later for not A, is very rare. If that happens, either your prior does not reflect your past evidence, your model does not fit the data, data is faulty, or you have bad luck. <br /><br />As I have seen, the main reason to use bayesian methods is not to inject prior knowledge into the model, but more technical. Hierarchical models have internal "priors", and on some other models maximum likelihood solutions are singular while the posterior still makes a lot of sense. In some cases, the model family is fine tuned with a prior. (e.g., L1 vs L2 vs Cauchy vs spike slab in linear regression). <br /><br />Often the priors are noninformative, carrying no domain knowledge. <br /><br />And note that your model family is also a prior, and a strong one!<br /><br />Mighty or Fred<br />Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-16612221.post-16700640836447129042013-02-07T03:34:57.984-05:002013-02-07T03:34:57.984-05:00I went through something like this about 20 years ...I went through something like this about 20 years ago. Some clever bunny discovered that you could "enhance" electron energy loss spectra using Bayesian statistics. This was great because EELS has problems with resolution and getting high quality spectra is tricky.<br /><br />I even saw said bunny give a talk about it at a conference. It looked complicated but wonderful and we all rushed off home to try it out for ourselves.<br /><br />Maybe I pressed the wrong button, but it seemed I could get any result I wanted from my data. Delta functions and everything. I came to the conclusion that I could significantly improve the resolution of my spectra, but only if I knew the answer precisely before I started.<br /><br />As far as I know, nobody uses this method nowadays for EELS, so I was possibly not alone.<br /><br />Danger MouseAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-16612221.post-47902490126196626672013-02-07T00:14:45.928-05:002013-02-07T00:14:45.928-05:00Bloody sock puppets!Bloody sock puppets!willardhttp://neverendingaudit.tumblr.comnoreply@blogger.com