ENBIS-16 in Sheffield

11 – 15 September 2016; Sheffield Abstract submission: 20 March – 4 July 2016

My abstracts


The following abstracts have been accepted for this event:

  • Modelling Bubbles and Crashes: Applications to Bitcoin and to Trading

    Authors: John Fry (Sheffield Business School)
    Primary area of focus / application: Other: Applied Statistics Meets Practical Statistics
    Secondary area of focus / application: Finance
    Keywords: Speculative bubbles, Econophysics, Bitcoin, Trading strategies
    Submitted at 5-Apr-2016 14:54 by John Fry
    Accepted (view paper)
    13-Sep-2016 10:10 Modelling Bubbles and Crashes: Applications to Bitcoin and to Trading
    We consider a stochastic or second-order extension of the seminal Johansen-Ledoit-Sornette model. During a bubble prices spike upwards. This behaviour can be described mathematically by a temporary increase in the drift function and a temporary decrease in the volatility function. The process is closely related to phase-transition phenomena in statistical physics and reflects an interesting overlap between physics and finance. Financially this reflects a collective market over-confidence and the fact that looking only at the historical price record is liable to under-estimate the true level of risk involved. We illustrate our model with financially interesting applications to Bitcoin and to the development of long-term trading strategies for stocks.
  • Assessing and Enhancing Conceptual Understanding: a Statistical and Educational Challenge

    Authors: Germana Trinchero (Ministry of Public Education, Turin), Ferdinando Arzarello (Department of Mathematic, University of Turin), Ron S. Kenett (Department of Mathematic, University of Turin), Ornella Robutti (Department of Mathematic, University of Turin), Paola Carante (Department of Mathematic, University of Turin)
    Primary area of focus / application: Education & Thinking
    Keywords: MERLO, Formative assessment, Teachers professional development, Mathematics education, Statistic education
    Submitted at 6-Apr-2016 13:41 by Germana Trinchero
    13-Sep-2016 10:10 Assessing and Enhancing Conceptual Understanding: a Statistical and Educational Challenge
    Assessing the level of conceptual understanding of students is both a statistical and a pedagogical challenge. In the context of Mathematics and Statistics, teachers need to be able to effectively convey concepts to students and feedback on their work is critical information. In this talk, we present experience gained in Italian secondary schools using a novel approach for assessing conceptual understanding called MERLO (Meaning Equivalence Reusable Learning Objects). MERLO is a didactic and methodological tool originally developed by Shafrir and Etkind to emphasize a pedagogical focus on conceptual understanding (Etkind et alii, 2010).
    What we call ‘MERLO pedagogy’ is composed of structured activities covering specific concepts within a discipline, through multi-semiotic representations in multiple sign systems as elements of items to be solved by the students, along with specific teaching practices and methodologies to be applied by the teacher with the students (Arzarello et alii, 2015). Each MERLO item includes five different statement representations that can share or not share the same meaning, being similar only in appearance. The innovation in MERLO consists of directly challenging students in discovering deep relations among different representations, and not in simply stating if they are true or false, or relate each other because they are similar in appearance (Arzarello et alii, in preparation). The teaching innovation is the design of these items by teacher educators, researchers, and teachers, according to the MERLO pedagogy.
    The talk will present an example based on MERLO student scores with results that can be generalized in the general context of statistical education. We will show how to compare performance of different classes, the advantage of group exercises and how one can identify the level of conceptual understanding of different concepts. This study shows how to meet the statistical challenge of measuring the level of understanding of students with implications on how Statistics can be taught.
  • Understanding the Inductance Part of the Lévy Generator: Some Applications

    Authors: Chris McCollin (Nottingham Trent University), Rainer Göb (University of Würzburg)
    Primary area of focus / application: Reliability
    Secondary area of focus / application: Process
    Keywords: Lévy process, Drainpipe theory, Japanese control chart, Langevin equation
    Submitted at 8-Apr-2016 12:38 by Chris McCollin
    Accepted (view paper)
    13-Sep-2016 14:50 Understanding the Inductance Part of the Lévy Generator: Some Applications
    Previous work by the authors on the Lévy Generator model (ENBIS 2013-15) have been concerned with the real part of the Lévy process and its relationship to the electronic–hydraulic analogy or drainpipe theory. The present work looks at the imaginary part of a specific Lévy Generator model and models it to nuclear pumps, the Japanese control chart and helicopter development data. The model identifies possible failure causing resonance, a changing frequency parameter which may indicate an impulse effect or a connection with the exponential part of the Langevin equation and some observations on processes and energy profiles which resemble chaotic systems in nature.
  • Likelihood-free Optimum Experimental Design: ABCD

    Authors: Werner G. Mueller (Johannes Kepler University)
    Primary area of focus / application: Other: ISBIS session
    Secondary area of focus / application: Design and analysis of experiments
    Keywords: Approximate Bayesian computation, Simulation-based design, Design criterion, Spatial extremes
    Submitted at 8-Apr-2016 16:13 by Werner G. Mueller
    Accepted (view paper)
    12-Sep-2016 14:00 Likelihood-free Optimum Experimental Design: ABCD
    We are concerned with improving data collecting schemes via methods of optimum experimental design, which can be applied in cases where the experimenter has at least partial control over the experimental conditions. Furthermore we focus on cases where a probability model for the investigated phenomenon is not easily available and the situation lends itself naturally to simulation-based approaches in conjunction with a recently popularized simulation technique called approximate Bayesian computing (ABC).

    The objective of optimum experimental design is to find the best possible configuration of factor settings with respect to a well-defined criterion or measure of information for a specific statistical model. In Bayesian experimental design, a prior distribution is attached to the parameters of the statistical model. This prior distribution reflects prior knowledge about the parameters of the model. In the Bayesian setting it is natural to average a criterion over the parameter values with respect to the prior distribution. In a decision-theoretic approach to experimental design the criterion of interest is computed for the posterior distribution of the parameters and then averaged over the marginal distribution of the data. The information criterion on the posterior distribution reflects some notion of learning from the observations.

    The computation of the expected criterion value can be a challenging task. Usually this involves the evaluation of integrals or sums. If the integrals are analytically intractable and numerical integration routines do not work, Monte Carlo simulation strategies can be applied in a framework of stochastic optimization. Some of our proposed methods will be based on simulation-based optimal design algorithms which utilize Markov chain Monte Carlo (MCMC) methods, but we intend to go beyond that class. Simulation-based methods make it possible to efficiently solve a wider range of problems for which standard methods cannot provide tractable solutions. In this presentation we outline potentials and limitations of ABC for design purposes, hence ABCD (D for design). Furthermore we will report details on an application for dealing with spatial extremes.
  • The Parameter Diagram as a DoE Planning Tool

    Authors: Matthew Barsalou (BorgWarner Turbo Systems Engineering GmbH)
    Primary area of focus / application: Design and analysis of experiments
    Keywords: Parameter diagram, Design of experiments, System understanding, Noise factors
    Submitted at 10-Apr-2016 14:13 by Matthew Barsalou
    Statisticians are often called upon to work together with Subject Matter Experts (SMEs) to perform Design of Experiments (DoEs). The statistician may have mastered DoE; however, the SME’s input may be critical in determining the correct factors, levels, and response variable of interest. The SME may be an engineer or even the machine operator responsible for the daily activities at the process that is being considered for a DoE. They may not understand what a DoE is or what is needed for a DoE. To facilitate DoE planning, a Parameter diagram (p-diagram) may be helpful. A p-diagram is not a new tool and it is often used in the automotive industry for the creation of Design Failure Modes and Effects Analysis. The use of a p-diagram as a DoE preparation tool, however, is a new application of the concept.

    A p-diagram is a graphical depiction of the inputs and outputs of a system. The inputs may be the main inputs such as raw material or a control signal in the form of voltage. There are also control factors, which are directly controlled such as when they are tested before hand to ensure they conform to requirements. Noise factors also influence a system, but these are uncontrolled such as weather or wear over time. Ideal functions and error states are also listed, with the ideal function being the desired output of the system and the errors states representing ways in which things could go wrong. Potential DoE factors can be drawn from the input and control factors while noise factors can be useful for identifying potential blocks. The response variable would either be the ideal function when it must be optimized or the error state when there is a condition that should be avoided. A p-diagram can be created in a presentation program or even on a whiteboard. However, SMEs are needed to provide the p-diagram inputs. These inputs are needed for a proper DoE and use of the p-diagram helps to ensure they are captured during DoE planning.

    This talk will describe the p-diagram and its application in DoE. Examples will be presented using actual DoEs from the literature. These case studies are the identification of the AA battery configuration with the longest life, improving the quality of a molded part, increasing the life of a molded tank deterrent device, and the optimization of a silver powder production process. After attending this talk, participants will be able to use a p-diagram for DoE planning.
  • Supersaturated Split-Plot Screening Experiments

    Authors: Emily Matthews (University of Southampton)
    Primary area of focus / application: Design and analysis of experiments
    Keywords: Screening, Design of Experiments, Split-plot, Supersaturated
    Submitted at 11-Apr-2016 12:13 by Emily Matthews
    Accepted (view paper)
    12-Sep-2016 10:40 Supersaturated Split-Plot Screening Experiments
    Screening is a key step in early industrial and scientific experimentation to identify those factors that have a substantive impact on the response. Practical screening experiments often have to be performed with limited resources and in the presence of restrictions in the randomisation of the design due to, for example, the need to incorporate hard-to-change factors via a split-plot design. Supersaturated designs, with fewer runs than the number of potential individual and joint factor effects, are now a common tool for screening experiments. We present a methodology that generalises this class of design to split-plot experiments. A linear mixed effects model is used to describe the response from such experiments, and methods for optimal design, model selection and variable-component estimation are developed and presented. An example from materials science and an industrial example from pharmaceutical science are used to demonstrate new approaches to both the design and analysis of such supersaturated split-plot experiments.