Ongoing To Do List

- Get broad thesis outline on paper/wiki - Find Berger and Selke paper and read it

Ongoing vague questions and what I want to read to answer/clarify them

- What is the link between bayesianism and a fully worked out philosophy of science? What, if anything, needs to be added to the bayesian framework for it to function as a ‘complete’ philosophy of science? How does methodology, confirmation theory etc all fit in with the bayesian framework? Read: philosophy of science - Cartwright/Hacking etc.. Finish H&U.. Earman - Will second-order probabilities cause any problems for practical use of bayesian methodologies? Berger - What are the problems of the choice of test statistics? Which are problems only for classical and which are problems for both classical and bayesian? H&U

Weekly Plan

w/b 18/9/06

- Read LANCET pre-publications on COX-2 inhibitors/diclofenac - Intro to Steve — ?inclusion LANCET and NEJM papers and other meta-analysis results

w/b 7/8/06

- Start thinking about Paper 2 - Mayo - Start sorting medical paper with simulations (with Steve) - Rofecoxibpaper And H&U

w/b 17/7/06

- Split rofecoxib paper into two - Paper 1 - focus on rofecoxib as case study; talk only about classical statistics and the way in which APPROVe was interpreted in practice - pretty much paper as it is currently be remove focus on Mayo. Target “Studies in Hist and Phil of Biol and Biomedical Sciences”. Complete ASAP. - Paper 2 - look to provide a reply to Mayo 2006 using studies like APPROVe as examples which raise questions which Mayo does not seem to be considering, ?or at least providing a strange answer to - highly statistically significant findings from under -powered studies. Target Br J Phil Sci then move down the list.

w/b 10/7/06

- wow, it has been a long time since i have done this - complete cox-2 draft (seriously) and send of within 2 weeks - talk to jason about upcomming rofexocib papers and correction in NEJM http://content.nejm.org/early_release/index.shtml#7-13-06 (?possible link with work melb people are doing) - Yeah. Isn’t it amazing that they can say “visual inspection of Figure 2 suggested that the Kaplan-Meier curves separated 18 months after randomization. However, the results of an overall test of the proportional-hazards assumption for the entire 36-month observation period did not reach statistical significance (P = 0.07).”? So whether two things (in the results, note, not in the population) differ depends on a significance test?! (CC Alison on this point.) Jason - sort Sydney - sort small paper with Steve on rofecoxib and decision risk with simulation data

w/b 19/3/06

- continue with cox-2 paper - send off finished jeffrey - continue with Mayo, have a crack at Berger & Selke

w/b 6/3/06

- start drafting cox-2 paper - read mayo - find woodward, lewis

w/b 6/2/06

- Complete Jeffrey r/v draft - Prepare rofecoxib presentation for popn health and pharmacy meeting - Write abstract for NMS

w/b 30/1/06

- Outline (at least) for the Jeffrey review. - Start thinking about some abstracts (and talk to MC, JG re thoughts about possible ‘medical’ conferences - NMS , disease-mongering, ?pharmacoepi) see Presentations Abstracts - Continue with Mayo and Hacking - Not sure I will get to it.. but.. new proposal outline

w/b 23/1/06

- the continuing joys of Jeffrey. start writing r/v. - look to get my head around the links between Jeffrey’s expectation discussion and H and U test statistic chapter - starting thinking through my claim that I can frame my key points on the rofecoxib story within a classical statistics framework

w/b 16/1/06

- Jeffrey: I am particularly interested in attempting to get some understanding of exchangability. - Still need to get back to test statistic stuff (some good overlap between H and U and Jeffrey here) - Selections from the generic list above

w/b 9/1/06

pretty much what is above… focussing on: - another couple of Jeffrey Chapters - Reply to “Why do we need large simple RCTs” - Test statistic stuff from [=HandU=]

w/b 3/1/06

(this is becoming a list of things I am procrastinating on - but i definately need to get the following done, if not this week then soon [see http://www-csli.stanford.edu/~john/procrastination.html — Jason]:) - Get somewhere with Jeffrey - or at least start listing what is stumping me - Finish off Hacking - Think about Cable Guy - Follow some of Goodmans references - Put together new [=PhD=] outline - Start rofecoxib paper - Refresh my reading list (linked with outline)

w/b 19/12/05

- Need to make some progress with Jeffrey’s Subjective Prob - Hacking

w/b 12/12/05

- Goodman papers - Jeffrey - get over Dorling; finish [=CH2=]/3 - Hacking (Intervening section - a couple of chapters)