WPMSIIP is an annual workshop on interval probability, or more generally imprecise probability, and related statistical inference. The explicit aim of the workshop is to stimulate discussion on ongoing research and open problems. Participants can of course present recent results if relevant for such discussions, but they can also just introduce open problems.
WPMSIIP 2019 returns to Durham, where the workshop series was started in 2008, and where it was also held in 2010 and 2016. It was further held in Munich (2009, 2012, 2015), Ljubljana (2011), Lugano (2013), Ghent (2014), Compiegne (2017) and Oviedo (2018).
The WPMSIIP workshop will be held at Durham University in 2019, between 9th and 12th September, with a networking activity (eg hiking) to take place on 13th.
Schedule
(Times of presentations are not pre-determined to provide flexibility to the workshop.)
Monday 9 September
Morning session 10:00-12:30
Computation: Organisers: Louis Aslett & Daniel Krpelik (Durham)
This session will investigate computational methods for assessing inferences from imprecise probability models. The aims are to discuss the current state-of-the-art, on-going research, and what obstacles stand in the way of application of imprecise probability models in broader practice.
-
Daniel Krpelik: “Sampling methods for imprecise probabilities”
-
Louis Aslett: “Rejecting rejection: thoughts on sampling for imprecise Bayesian models”
-
Tathagata Basu: “Sensitivity Analysis of Adaptive LASSO”
Afternoon session 13:30-16:00
Decision making: Organisers: Camila Caiado & Matthias C. M. Troffaes (Durham)
The main goal of the session is to generate discussions on open problems that relate to decision making under severe uncertainty, modelled via imprecise probability, or via other frameworks targetting severe uncertainty. Potential topics include algorithms, choice functions, utility theory, axiomatic frameworks, and so on. We also welcome applications that involve decision making but where probabilities are hard to fully quantify.
-
Patrick Schwaferts: “Bayesian decisions using regions of practical equivalence (ROPE) and imprecise loss functions”
-
Matthias C. M. Troffaes: “Algorithms and benchmarking for imprecise decision criteria”
-
Camila Caiado: “Evaluating betting odds and free coupons using desirability”
Tuesday 10 September
Morning session 10:00-12:30
Decision making: Organisers: Camila Caiado & Matthias C. M. Troffaes (Durham)
The main goal of the session is to generate discussions on open problems that relate to decision making under severe uncertainty, modelled via imprecise probability, or via other frameworks targetting severe uncertainty. Potential topics include algorithms, choice functions, utility theory, axiomatic frameworks, and so on. We also welcome applications that involve decision making but where probabilities are hard to fully quantify.
-
Michael Goldstein: “Bayes linear imprecision and decision support for complex computer models”
-
Hailiang Du: “Bayes linear optimization for energy system planning”
-
Henna Bains: “Methods to communicate decision making using imprecise probabilities with a wider audience”
Afternoon session 13:30-16:00
Risk and Reliability: Organiser: Frank Coolen (Durham)
Risk and reliability are areas with many interesting practical problems and a wide range of challenges for quantitative methods. Information may be scarce, of low quality or only available from related settings. Decisions may require to take preferences from many people or no specific people into account. Imprecision of probabilities and utilities may provide suitable tools in these areas, many challenges remain.
-
Frank Coolen and Sultan Albalwy: “Accelerated life testing”
-
Zarif Zaman: “Analytical Probability Propagation Method for Reliability Analysis with Interval Probabilities”
-
Jinlei Qin: “Reliability Analysis on the Multi-state system”
-
Ahmad Albaity: “Proportional hazards model”
Wednesday 11 September
Statistical Inference: Organiser: Tahani Coolen-Maturi & Frank Coolen (Durham)
Imprecision can be included in statistical methods in a variety of ways. This session is aimed at reasons for doing so and suitable methods, in particular with focus on the benefits of imprecise statistical methods compared to standard approaches.
Morning session 10:00-12:30
-
Andrea Simkus, Fatimah Alghamdi and Frank Coolen: “Reproducibility of tests”
-
Abdulmajeed Alharbi and Masad Alrasheedi: “Topics in Classification”
-
Tahani Coolen-Maturi: “The ordering of future observations from multiple groups”
Afternoon session 13:30-16:00
-
Georg Schollmeyer: “Some thoughts about using imprecise statistical characteristics as predicates/statistical invariants in classification and hypothesis testing”
-
Asamh Alluhayb: “Smoothed Bootstrap for right-censored data”
-
Kholood Alyazidi: “Inventory decisions”
Thursday 12 September
Statistical Inference: Organiser: Tahani Coolen-Maturi & Frank Coolen (Durham)
Imprecision can be included in statistical methods in a variety of ways. This session is aimed at reasons for doing so and suitable methods, in particular with focus on the benefits of imprecise statistical methods compared to standard approaches.
Morning session 10:00-12:00
-
Ali Manashi: “NPI for mutiple future observations with right-censored data”
-
Ahmad Albaity and Frank Coolen: “Most-likely-data estimation”
Lunch 12:00-14:00
Lunch in the city center
Afternoon session 14:00-16:00
General discussions and Arising Topics
Friday 13 September
An excursion type activity.
Fees
There will be NO FEES to participate in the meeting. Participants should book accommodation themselves and also pay for meals and refreshments, although we aim to provide refreshments during breaks within the sessions. We will aim to organize a joint dinner on Wednesday or Thursday evening, and will contact participants about this in advance of the meeting.
Registration
Please contact Tahani Coolen-Maturi (tahani.maturi@durham.ac.uk) by 15 August to register for the meeting.
We hope that all participants give at least one presentation or introduce a discussion on a topic of interest. Participants are also very welcome to arrange a session of multiple presentations or discussions. Please contact Tahani (tahani.maturi@durham.ac.uk) as soon as possible with any plans or suggestions for such contributions.
Venue
- Monday-Wednesday talks: room ES236 (2nd floor Earth Sciences, Arthur Holmes building opposite Maths)
- Thursday: room CM301 (Maths department)
- Lunches on Monday-Wednesday are in CM211 (Maths department)
Accommodation
We recommend participants to book accommodation at Durham Colleges, in particular at St Cuthbert’s Society where we have reserved some rooms. Durham City also has several hotels at walking distance from the venue.
If you would like to book at St Cuthbert’s Society, please click here.
Local Organisation
Louis Aslett (Durham University)
Tathagata Basu (Durham)
Frank Coolen (Durham University)
Tahani Coolen-Maturi (Durham University)
Daniel Krpelik (Durham University)
For any questions feel free to contact Prof Frank Coolen (frank.coolen@durham.ac.uk) or Dr Tahani Coolen-Maturi (tahani.maturi@durham.ac.uk)
Getting to Durham
Details about how to get to Durham are available from the university website here. The city lies on the Eastern mainline railway served by LNER, CrossCountry Trains and TransPennine Express. It is also just 10 minutes off the A1(M) motorway. From further afield, flights are available to Newcastle International Airport.
Key locations
Click the markers in the Google Map below for details about key locations for the workshop:
Links
The Society for Imprecise Probability: Theories and Applications