apts .ac.uk Week 3: University of Southampton Academy for PhD Training in Statistics

advertisement
apts.ac.uk
Academy for PhD Training in Statistics
Week 3: University of Southampton
11th - 15th July, 2011
Welcome to Southampton!
The City of Southampton: The city has numerous parks and open spaces, including Southampton Common, situated near our Highfield Campus. The waterfront location also provides countless opportunities for sport and leisure with a wide range of water sports to appeal to all abilities.
Southampton is home to one of the UK’s top 10 shopping centres, West Quay, and the city
centre and waterfront marina both offer a range of independent shops, as well as restaurants,
cafeĢs, bars, pubs and clubs. With cinemas, theatres, galleries and museums, the city offers a
rich assortment of cultural attractions.
Surroundings: Winchester, once the capital of England, has a rich cultural heritage and lively
city atmosphere with galleries, museums, theatres, cinema and arts centre, as well as a wide
variety of shops, pubs and restaurants. Furthermore, the New Forest National Park is the
largest unenclosed area in southern England, where ponies, deer and cattle graze freely, in an
environment that remains unchanged by the modern world. It stretches for 145 square miles
(375km2 ), with open heaths and beautiful forest landscapes.
Workshop registration: Check-in for accommodation, and registration for the APTS week
will be at the APTS registration desk, situated in the level 1 foyer of the Murray Building
(building 58 on the Highfield Campus map), between 11.15 and 14.15 on Monday July 11th.
You will receive your welcome pack and badge from the registration desk. Please wear your
badge at all times. This will help with security and also help you identify fellow participants.
IT: Computing facilities are available from 09.00 - 18.00 in the foyer of the Murray Building.
The login details will be supplied on the first day of the workshop in your welcome pack.
Car parking: Workshop participants are able to use the Halls of Residence car park but spaces
are limited. Car parking in the Halls of Residence is free of charge during the APTS week. Car
parking is by permit only (could you please let us know via email by 30th June if you require
a permit).
Messages: The telephone number for colleagues or family to leave an urgent message for you
during office hours is +442380 594548. For emergency messages outside these times, please call
the main University Switchboard on +442380 595000.
Other campus facilities: Facilities at the Highfield Campus include a University shop, Post
Office and bookshop. There are Lloyds TSB, HSBC, Santander, Natwest and Barclays Banks
on campus with cash dispensers. You will be able to use the state-of-the-art Jubilee Sports
Centre (building 18 on the Highfield Campus map), with a payment of £10 for unlimited use of
all sports facilities during your stay (you will be issued with a mini-card which will be used to
access the sports building). Please note that due to the short duration of the workshop there is
not enough time for induction to the fitness suite facilities. Thus the fitness suite can be used
only by participants that have experience with similar facilities.
Accommodation information
Your room: Workshop participants requiring accommodation have been booked at rooms in
Highfield Hall. These will be available from 3.00pm on Monday 11th July. Secure luggage
storage will be available close to registration in the Murray Building (Building 58) for the
afternoon of 11th July and again on the morning of Friday 15th July.
Reception/Keys: You should collect your keys from the APTS reception desk on Monday
11th July, 11.15am – 2.15pm in the foyer of the Murray Building. Please vacate your room by
9.00am on the day of departure and bring your keys to the APTS reception desk in the foyer
of the Murray Building.
Internet access: All the rooms on campus are networked for free internet access, although you
will need to provide your own ethernet cable. Alternatively, PCs with free internet and e-mail
access are available on the Highfield Campus in the foyer of the Murray Building. Wireless
internet access is available in the Murray building using the Southampton login you will be
provided with, or by using eduroam.
Meals:
Breakfast: Highfield Dining Room (Dining Hall at Highfield Hall on the map below) will provide
a full English or continental breakfast from 08.00 am – 08.30 am, each morning of your stay.
Lunch: There will be a finger buffet on the Monday and the Friday in the foyer of the Murray
Building. On the remaining three days lunch will be served in the University Social Centre
(building 38 on the Highfield Campus map).
Evening: Evening meals will be provided in the Highfield Dining Room (Monday to Wednesday)
serving a three course cafeteria style dinner. The Academy Dinner on Thursday will be held in
the University Social Centre with a bar which serves a selection of beers, wines, spirits and soft
drinks. Dinner will be ready to be served at 7.30pm. The bar will be open from 7.00pm.
Emergency information
First Aid and accident reporting:
In the event of an accident that needs medical attention the nearest first aider should be
contacted. The names and locations of trained first aiders are displayed on green and white
signs throughout each building. In addition, all Security staff are trained first aiders and can
be contacted on 023 8059 331.
In the event of a serious accident, the ambulance service should be contacted immediately by
dialling 91-999 before calling a first-aider and notifying the Central Control Room (CCR) on
extension 3311. The CCR non-emergency contact number is 22811.
Fire safety and emergency procedures:
Action in the event of fire in other building: If you notice a fire you should immediately raise
the alarm by breaking the glass of the nearest manual fire alarm call point. This can be done
using your elbow or shoe. The alarm is a continuous bell.
Calling the Fire Service: The person raising the alarm should ensure that the Fire Service is
summoned by either reporting to someone in authority, or by dialling 91-999 from a telephone
remote from where the alarms are sounding. Following this, the University Control Centre
Room (CCR) should be alerted by dialling 3311.
On hearing the continuous ringing of fire bells:
Stop what you are doing.
Leave by the nearest Fire Exit.
Walk calmly, do not run.
Do not stop to collect personal belongings.
Make your way to the nearest evacuation point, standing well clear of the building.
Do not re-enter the building until told to do so by the Fire Service or University Security staff.
Highfield Campus map
APTS Timetable
Monday
11th Jul
09.15 – 10.45
Tuesday
12th Jul
Wednesday
13th Jul
Thursday
14th Jul
Friday
15th Jul
Applied
Stochastic
Processes
Applied
Stochastic
Processes
Applied
Stochastic
Processes
Applied
Stochastic
Processes
10.45 – 11.15
11.15 – 12.45
Tea and Coffee
Registration
Computer
Intensive
Statistics
12.45 – 14.00
Welcome
14.15 – 15.15
Applied
Stochastic
Processes
(14.15 –
15.45)
15.15 – 16.00
Applied
Stochastic
Processes
Applied
Stochastic
Processes
Special
Lecture
Prof. Sir David
Cox
Tea and Coffee
Computer
Intensive
Statistics
(16.15 –
17.45)
19.00 – 20.00
Evening
Computer
Intensive
Statistics
Lunch
14.00 – 14.15
16.00 – 17.00
Computer
Intensive
Statistics
Computer
Intensive
Statistics
(Computer
lab)
Computer
Intensive
Statistics
(Computer
lab)
Computer
Intensive
Statistics
(Computer
lab)
Free Evening
Academy
dinner
(19.00 – )
Dinner
RSS Wine
Reception
(20.00 – )
Free Evening
(Please see the accompanying notes on the following page.)
Computer
Intensive
Statistics
Timetable notes
• Locations: All the lectures will take place in the Murray Lecture Theatre (Room 1067)
in Building 58 (Murray). The computer labs will take place in rooms 1061 and 1057
(laptops only) of Building 44 (Shackleton).
• Laboratory sessions: You are encouraged to bring your own laptop to work on during
these sessions. Detailed information about software requirements and configuration can
be found at: http://www.stats.ox.ac.uk/~ripley/APTS2011/.
• Special lecture: We’re very pleased to welcome Professor Sir David Cox (http://www.
stats.ox.ac.uk/people/associate_staff/david_cox), an eminent statistician and excellent speaker, to deliver a special lecture.
• RSS wine reception: This will take place in Highfield Hall.
• Tea and coffee: Tea and coffee will be served in the foyer of the Murray Building.
• Lunches: On Monday and Friday there will be buffet-style lunch in the foyer of the
Murray Building. On the three other days lunch will be served in the University Social
Centre (building 38 on the Highfield Campus map).
• Dinners: All participants are booked for dinner at Highfield Dining Hall on MondayWednesday evenings.
• Academy dinner: This will take place at University Social Centre. The bar will be open
from 7.00pm and dinner will be served at 7.30pm. Smart-casual dress is recommended.
Module Details
Computationally Intensive Statistical Methods
Module leader: B.D. Ripley
Aim: This module will introduce various computationally-intensive methods and their background theory, including material on simulation-based approaches such as Markov-chain Monte
Carlo (MCMC) and the bootstrap, and on strategies for handling large datasets. The different
methods will be illustrated by applications.
Learning outcomes: After taking this module, students will have a working appreciation of
MCMC, the bootstrap and other simulation-based methods and of their limitations, and have
some experience of implementing them for simple examples. Students will also have gained
an appreciation of the difficulties of handling very large datasets and of some approaches to
overcoming them.
Prerequisites: Preparation for this module should include a review of:
• relevant basic material on statistical modelling (for which the earlier APTS module ‘Statistical Modelling’ would be advantageous);
• basic Markov chains (as for the ‘Applied Stochastic Processes’ module). Basic knowledge
of programming in a high-level language such as R will be assumed, and R will be used
for case studies and exercises.
Topics:
• Overview of simulation-based inference; Monte Carlo testing.
• Basic theory of bootstrap methods; practical considerations; limitations.
• Basic theory of MCMC; types of MCMC samplers; assessment of convergence/mixing;
other practical considerations; case studies.
• Strategies for dealing with large datasets: use of database management systems, multipass
algorithms, subsampling, distributed computing. A case study, e.g. generalized linear
models.
Assessment: Analysis of a data set using computationally-intensive methods. This could be a
data set related to the student’s research project, or an example specified by the module leader.
Applied Stochastic Processes
Module leaders: S. B. Connor & W.S. Kendall
Aims: This module will introduce students to two important notions in stochastic processes
reversibility and martingales identifying the basic ideas, outlining the main results and giving
a flavour of some of the important ways in which these notions are used in statistics.
Learning outcomes: A student successfully completing this module will be able to:
• describe and calculate with the notion of a reversible Markov chain, both in discrete and
continuous time;
• describe the basic properties of discrete-parameter martingales and check whether the
martingale property holds;
• recall and apply significant concepts from martingale theory (indicative list: optional
stopping, martingale convergence);
• explain how to use Foster-Lyapunov criteria to establish recurrence and speed of convergence to equilibrium for Markov chains.
Prerequisites: Preparation for this module should include a review of the basic theory and concepts of Markov chains as examples of simple stochastic processes (transition and rate matrices,
irreducibility and aperiodicity, equilibrium equations and results on convergence to equilibrium),
and with the definition and basic properties of the Poisson process (as an example of a simple
counting process).
Topics:
• Reversibility of Markov chains in both discrete and continuous time, computation of equilibrium distributions for such chains, application to important examples.
• Discrete time martingales, examples, application, super-martingales, sub-martingales.
• Stopping times, statements and applications of optional stopping theorem, martingale
convergence theorem.
• Recurrence and rates of convergence for Markov chains, application to important examples.
• Statements and applications of Foster-Lyapunov criteria, viewed using the language of
martingales.
• Statistical applications and relevance (highlighted where appropriate throughout).
Assessment: One of
• Read an appropriately chosen paper (either specified or from a specified list), and identify
some of the main stochastic process models and results referred to in the paper, describe
their significance and use in the paper, and relate them to the material covered in the
module.
• Complete appropriate exercises that are simple developments or extensions of aspects of
the results in the module (that may be related in some way to models and results arising
in the paper).
Notes
Download