ninepointeightone.net


Statistics Seminar - 17/03/2008

4.00pm Monday, Mathematical Institute Theatre D

Dr. Vincent Macaulay, University of Glasgow

“Out of Africa: moving on “

A recent dispersal of anatomically modern humans out of Africa is now widely accepted, partly due to the pioneering work of members of the Wilson lab examining genetic variation in the maternally-transmitted genome mitochondrial DNA (mtDNA). Subsequent work on other parts of the genome has in general supported this conclusion (although not unequivocally). In this presentation, I will review the evidence that has led to this position, but go on to show that the evidence from mtDNA is not yet exhausted. I will report recent work of myself and others that has attempted to tease out, using a little phylogenetics and probability, more of the characteristics of the process by which modern humans came out of Africa. I shall in particular address the route taken across Eurasia, which is hotly disputed. The mtDNA variation in isolated “relict” populations in southeast Asia supports the view that there was only a single dispersal from Africa, most probably via a southern coastal route, through India and onward into southeast Asia and Australasia. In this picture, there was an early offshoot, leading ultimately to the settlement of the Near East and Europe, but the main dispersal from India to Australia ~65,000 years ago was rapid, most probably taking only a few thousand years.

2008-03-17 @ 21:33:52

IPA talk 11/03/08

7pm, Holy Trinity Church, St Andrews.

Dr Hans Blix

“Time for a Revival of Disarmament?”

Best quotations:
A wise American friend of mine said: “I always trust those who seek the truth and mistrust those who claim to know it.”
During my work as a UN weapons inspector I was bugged by the US government, if only they’d paid attention to what I said(!)

2008-03-11 @ 23:49:13

CREEM Seminar - 27/02/08

4.00pm Wednesday, CREEM
Undecaffeinated coffee and tea from 3.45pm

Tiago A. Marques (Joint work with Len Thomas)

“DECAF - Overview and initial results”

The goal of this talk is to describe the project I am currently working on as part of my post-doc and some of the results obtained so far. DECAF, the project acronym, stands for “Density Estimation for Cetaceans from passive Acoustic Fixed sensors”. It is a 3 year project, built around 4 case studies: sperm whales (2 case studies) at AUTEC, beaked whales at AUTEC and humpback whales at PMRF. For all of them, the ultimate objective is density estimation.

The background for the use of passive acoustics will be discussed, contrasting this approach with visual line surveys.

Until now most of the work was related to the beaked whales case study. We started by evaluating a method previously used at AUTEC to estimate beaked whale density. This was a group counting algorithm, with several potential problems related to some arbitrary choices and a statistical artifact: these are briefly reviewed. Building on that, we have evolved into looking at two different approaches: (1) dive counting and (2) cue counting. Both of these will be described. A likelihood based method to estimate a whale location was attempted, but just recently we realized that it was not adequate, and improvements to this will be attempted next. Further, data analysis to obtain relevant parameters to carry simulation of plausible data sets to evaluate the “dive counting” approach have also been attempted, with interesting preliminary results.

Recently we have started looking into the sperm whales case study, and some general ideas about the way to address this case study will also be presented.

2008-02-27 @ 23:53:05

CREEM Seminar 13/02/2008

4.00pm Wednesday, CREEM

Ali Johnstone, University of Cambridge

“Detectability of breeding birds”

The aim of this work is to estimate what proportion of birds breed at various ages. Existing methods such as Pradel cannot be used, as various assumptions are violated. Therefore, I’m trying to estimate the number of pairs which nest in a given breeding season, but are not detected, using closed population mark-recapture methods within the breeding season. I will compare the initial results, and hopefully spark discussion on other possible approaches to this problem.

2008-02-14 @ 19:43:14

Accepted

David Lawrence Miller:
Congratulations, your submission A mixture model approach to distance sampling detection functions has been accepted for presentation as a poster at International Statistical Ecology Conference at St. Andrews.

2008-01-31 @ 16:16:25

Copson Lecture 10/12/2007

Monday 10th December 5.15 pm, Lecture theatre A, Physics Building, University of St Andrews

Dr John Haigh, University of Sussex

“How Likely is That?”

This lecture was established in memory of Edward Copson, Regius Professor of Mathematics in the University of St Andrews from 1950-1969, and aims to bring to St Andrews a distinguished mathematician with the reputation of being able to communicate mathematical ideas to a non-specialist audience.

Dr Haigh is an expert on probability who has written a popular book on the subject “Taking chances: winning with probability” and has given one of a series of popular lectures for the London Mathematical Society. Jointly with fellow sports lover Rob Eastaway he has also written “Beating the Odds; the hidden mathematics of sport”. He sums up his lecture entitled “How likely is that” as follows. “Answers to questions about probability are often surprising and may even seem paradoxical. But a logical approach shows why these answers arise”.

2007-12-10 @ 20:04:22

CREEM seminar 29/11/07

4.00pm Wednesday. Coffee from 3.45 pm, CREEM

Christian Ewald, Economics, University of St Andrews

“The conflict over parental care: A game theoretic analysis on aspects such as cooperation, benefit and welfare for the young”

In this talk we address various issues which arise in the classical parental care conflict, which have, should or could be modeled with game theoretic models. After a brief summary of fundamental concepts in game theory, aspects such as cooperation, benefit and welfare consequences for the young will be put into a game theoretic context and various models will be discussed. These models have been developed in joint work with John McNamara and Alasdair Houston.

2007-11-28 @ 17:08:33

seminars and road trips

Yes, I am still alive. For those who are interested in my latest geekery they can check out my journey (with Fearghas and his iPhone) from St Andrews to London last weekend. The moblog of the roadtrip is here.

Additionally, I’ve decided to start posting the abstracts from seminars I’ve been to here (as much for me to remember as any other reason.)

2007-11-14 @ 13:10:11

Train blogging…

So, I’m writing this from the 0920 GNER from Leuchars to London Kings Cross. Fortunately due to the excellent Google Gears, I can read my RSS feeds and catch up with the 60ish unread items I currently have sitting in my Google Reader.

I’ve recently subscribed to two new feeds, both of which I find excellent. The first of these is 43Folders, which is written for GTD addicts (like Merlin Mann, the author of the blog) and wannabes (like me.) I’d been aware of 43folders for a while, but not really paid it any attention. However I recently watched a Google Tech Talk from Merlin Mann about “Inbox Zero” which is a GTD technique to minimize the amount of time one spends in their mail reader. Having had this problem for some time, I found his tips really useful, although his comparison of the Blackberry to a spontaneously combusting cat a little tenuous (watch it, you’ll understand.)

My other recent feed acquisition is Coding Horror, which has some neat tips and tricks for those doing webdev. I found it while looking for a good explanation of SQL JOINs (which I found.)

So, a few items in, I’ve stumbled across Dave Gorman’s latest entry. He offers an an excellent solution to both the current “phone in scandal” problem and the postal strikes.

From Kernel Trap an excellent initiative by the Linux kernel team to have a list of projects available for (CS) students. This is a great idea, no more lame “I’m going to build an online shopping system” or “the world really needs another CMS”-type projects. Students can feel like they are doing something useful and the kernel team get some of the work done for them.

Andrew Gelman received an e-mail about bad analysis, that didn’t really concern me, but the last two lines are absolutely true. He also has a nice article about the difference between an article and a web page in academic research. I like the idea of publishing your research as a web page and getting it out to as wide an audience as possible. This seems like what research should really be about. Not just getting your paper into JASA or Biomerika, but actually getting the guy on the street to understand what scientists do all day.This does of course mean more work for the authors, but why be a researcher if you are not trying to tell people something they didn’t know about the world.

Stephan Fry’s latst blog entry is on fame. It is, as usual, lengthy but ultimately satisfying. My favorite paragraph being:

Is it fun? Or, as student journalists always ask, what’s it like? ‘What’s it like working with Natalie Portman, what’s it like doing QI, what’s it like being famous?’ I don’t know what it is like. What is being English like? What is wearing a hat like? What’s eating Thai red curry like? I don’t believe that I can answer any question formulated that way. So, student journalists, tyro profilers and rooky reporters out there, seriously, quite seriously never ask a ‘what’s it like’ question, it instantly reveals your crapness. I used to try getting surreal when asked the question and say things like ‘being famous is like wearing blue pyjamas at the opera. It’s like kissing Neil Young, but only on Wednesdays. It’s like a silver disc gummed to the ear of a wolverine. It’s like licking crumbs from the belly of a waitress called Eileen. It’s like lemon polenta cake but slightly wider. It’s like moonrise on the planet Posker.’ I mean honestly. What’s it like?? Stop it at once.

Finally, a quick word about what else I’ve been up to. Mostly this week I’ve been attempting to do some work on my dissertation. This has been pretty tough, all in all. I really enjoy reading papers and trying to work out what’s going on, but sometimes it’s extremely frustrating when you reach an impasse and work stops for half a day until you figure out how you were being stupid this time.

While I’ve been working, I’ve been listening to a few new albums I’ve acquired over the past week or so. My dad sent me Amy Winehouse’s album, “Back to Black” which I realise I should probably slate but I have to say it’s great. A really good pop record, not much more than that, but really great at what it does. He also sent me a Freestylers double CD, “Different Story, Vol. 1″ which is a whole lot of drum’n'bass. It’s great to code to. Finally, I also downloaded the new Radiohead album. A lot has been said online about it and about how it is going to change the business model for online music. I’m not going to echo all of these comments here, I’ll just say it’s an excellent record. Maybe not as good as Thom Yorke’s “Eraser” from last year, but still excellent.

At this point the train pulls into Edinburgh Waverley, the train fills with the smell of brewery and I think I should probably do some real work.

2007-10-22 @ 16:14:32

hrmmm

Having just installed Ubuntu on a CREEM laptop, I find that fdisk runs claiming to have not been run in 49710 days, which is over 130 years… Interesting…

2007-09-26 @ 19:18:28

What I’m up to and some links…

So, I’ve been working on some top secret programming at the moment. As a teaser I’ll let you know that it involves Python, a tube map and problems in P.

On the links front: a nice user script to take Facebook events and import them into Google Calendar, a nice story about the St Andrews philosophy football match (see the section The Metaphysical Philosophers) and finally a cool app to use your GMail account as a filesystem in Mac OS.

More to come on the project at a later date.

2007-09-24 @ 12:50:53

3601 words…

… and my Wildlife Population Assessment project is done! Yay!

2007-05-02 @ 08:47:39

long time no blog… (#213)

It has been a very very long time since I last blogged, I’ve done a lot; for a start I went to Krakow which was a lot of fun. Today hall got me a new desk, which is exciting. It’s very very tall so I guess this is how people who are a normal height feel when they use a desk. I also, as usual have a lot of linkage…

If you want to cheat on your partner, why not CheatNeutral, it’s just like carbon neutralisation and just as moral! Lifehacker offers a way to calculate caffeine intake (needless to say, mine was way too high.) There is a very exciting website called LibriVox which offers audiobooks (in mp3 and ogg) for your iPod pleasure. It came to my attention because they have Darwin’s Origin of Species (via boingboing.) Finally, the classic problem is solved: Estimating the Airspeed Velocity of an Unladen Swallow:Hashing out the classic question with Strouhal numbers and simplified flight waveforms. Nice.

2007-04-26 @ 17:09:37

back in the lab

What have I been up to recently? Well, a lot of practicals for wildlife population assessment, and an essay for Adv. Bayesian Inference for a start. The former was not very fun but the second was pretty interesting although I’m not sure if I have any writing skills whatsoever (as those who read this blog will be able to confirm.)

On Wednesday it was “Pi Day” (3.14) I was not involved in the silliness of drinking wine in the physics common room before mid-day, my opinion is better summed up with the power of Dinosaur Comics.

In other linkage: MR on fairtrade, Steven Levitt gets annoyed with airlines (Freakonomics blog) and Ben Goldacre get angry about the global warming “documentary” which was on Channel 4 last week.

Finally, lightbluetouchpaper has a good article on how to write an abstract.

I think that’s all for now…

2007-03-16 @ 09:50:20

back from the grave assignments

If I were a Springer-Verlag Graduate Text in Mathematics, I would be J.L. Doob’s Measure Theory.

I am different from other books on measure theory in that I accept probability theory as an essential part of measure theory. This means that many examples are taken from probability; that probabilistic concepts such as independence, Markov processes, and conditional expectations are integrated into me rather than being relegated to an appendix; that more attention is paid to the role of algebras than is customary; and that the metric defining the distance between sets as the measure of their symmetric difference is exploited more than is customary.

Which Springer GTM would you be? The Springer GTM Test

2007-03-13 @ 22:40:07

20 queries. 0.961 seconds.
Powered by WordPress

Close
E-mail It