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:

  • Selection of a Subgrid from a Spatial Monitoring Process

    Authors: Riccardo Borgoni (Dipartimento di Economia, Metodi Quantitativi e Strategie d’Impresa, Università Bicocca, Milan), Diego Zappa (Department of Statistical Sciences, Università Cattolica del Sacro Cuore, Milan)
    Primary area of focus / application: Process
    Secondary area of focus / application: Other: Italian Session
    Keywords: Subgrid, Dual mean-variance response surface, Case studies, Spatial monitoring process
    Submitted at 11-Jul-2016 14:02 by Riccardo Borgoni
    13-Sep-2016 12:40 Selection of a Subgrid from a Spatial Monitoring Process
    Motivated by a real semiconductor problem, we propose a method to extract a subgrid from a given monitoring one, that may flexibly depend upon both/either the spatial representativeness and/or by expert knowledge of spatial weight to be assigned to those areas where production may need greater accuracy. Furthermore, conditionally upon the availability of experimental data collected using the initial grid we check the loss of accuracy by estimating a dual mean-variance response surface on the reduced grid. Joining the latter information and the criteria used to select the subgrid, we can provide additional guidelines on how to fine-tune the selection of the starting grid. Case studies are used to show the effectiveness of the procedure.
  • Local Gaussian Process Approximation for Large Computer Experiments

    Authors: Robert B. Gramacy (Virginia Tech)
    Primary area of focus / application: Modelling
    Secondary area of focus / application: Other: US Session
    Keywords: Large computer experiments, Gaussian process, Cluster computing, Big Data
    Submitted at 12-Jul-2016 17:33 by Robert Gramacy
    12-Sep-2016 15:00 Local Gaussian Process Approximation for Large Computer Experiments
    We provide a new approach to approximate emulation of large computer experiments. By focusing expressly on desirable properties of the predictive equations, we derive a family of local sequential design schemes that dynamically define the support of a Gaussian process predictor based on a local subset of the data. We further derive expressions for fast sequential updating of all needed quantities as the local designs are built-up iteratively. Then we show how independent application of our local design strategy across the elements of a vast predictive grid facilitates a trivially parallel implementation. The end result is a global predictor able to take advantage of modern multicore architectures, GPUs, and cluster computing, while at the same time allowing for a non-stationary modeling feature as a bonus. We demonstrate our method on examples utilizing designs sized in the tens of thousands to over a million data points. Comparisons are made to the method of compactly supported covariances, and we present applications to computer model calibration of a radiative shock and the calculation of satellite drag.
  • Monitoring and Improving Surgical Outcome Quality

    Authors: Stefan Steiner (Dept. of Statistics and Actuarial Science, University of Waterloo)
    Primary area of focus / application: Quality
    Keywords: Monitoring, Variable life-adjusted display, Risk-adjusted cumulative sum, Weighted estimating equation
    Submitted at 15-Jul-2016 14:13 by Stefan Steiner
    Accepted (view paper)
    13-Sep-2016 12:40 Monitoring and Improving Surgical Outcome Quality
    In this talk I will provide an introduction and overview of methods for monitoring surgical quality. In contrast to the monitoring of industrial processes, in this context I will motivate the need for risk adjustment and provide some examples. Next, I will describe and contrast previously proposed methods for surgical outcome monitoring, such as the variable life-adjusted display (VLAD), the risk-adjusted cumulative sum (RA-CUSUM) and the weighted estimating equation (WEE) approach. Applications to monitoring a single surgeon or surgical centre and comparisons involving multiple surgeons, centres or hospitals and implementation issues will also be briefly discussed. As time allows I will also introduce the National Surgical Quality Improvement Program (NSQIP) run by the American College of Surgeons.
  • Public and Private Capital in Indian Agriculture: An Analysis of Causality

    Authors: Anita Kumari (FASAL Unit at Institute of Economic Growth, Delhi), Nilabja Ghosh (FASAL Unit at Institute of Economic Growth, Delhi), M. Rajeshwor (FASAL Unit at Institute of Economic Growth, Delhi)
    Primary area of focus / application: Economics
    Keywords: Time series analysis, Granger causality, Indian agriculture, Public capital, Private capital
    Submitted at 22-Jul-2016 14:37 by Anita Kumari Gupta
    13-Sep-2016 14:50 Public and Private Capital in Indian Agriculture: An Analysis of Causality
    Capital enhances the productive capabilities of other inputs of production. There has always been a discussion whether there has been any causative relationship between public capital and private capital especially in Indian agriculture. In the study, therefore, a time series analysis was conducted to find out if the two types of capital, public and private, used for agriculture are mutually causative. An analysis was also done to find out if there has been a co-integration relationship between the two types of capital. The method of Granger Causality revealed that there exists causality between public and private capital including its components as well in both the directions. Both, public capital and private capital including its components have also been found to be co-integrated. Hence, an important inference drawn from the analysis for the Indian agriculture has been that both types of capital are needed as there has been no crowding out between public and private investment but these mutually influence each other having a complimentary relationship. Public investment on capital formation may encourage farmers to make investments and is also sensitive to existing development status of the farming sector.
  • Feedback Adjustment for Machining Processes

    Authors: Keith Harris (University of Sheffield), Eleanor Stillman (University of Sheffield), Kostas Triantafyllopoulos (University of Sheffield)
    Primary area of focus / application: Modelling
    Keywords: Manufacturing, EWMA, Feedback adjustment, Effect of misspecification
    Submitted at 28-Jul-2016 15:24 by Keith Harris
    14-Sep-2016 09:40 Feedback Adjustment for Machining Processes
    Within the field of manufacturing, there is tremendous interest in developing fully automated machining processes in order to minimize the cost of production and to standardize the quality of the finished product. To achieve this goal, the development of more sophisticated statistical methods of feedback adjustment is paramount. In this presentation, we will present results from a simulation study motivated by data from real machining experiments and designed to investigate different intervention strategies, including the popular EWMA (Exponentially Weighted Moving Average) scheme. We will also explore the effect of misspecifying the process gain and treating the process gain as a random variable.
  • Expediting Statistical Innovation Through Pre-Competitive Industrial and Academic Networks

    Authors: Luc Bijnens (Hasselt University)
    Primary area of focus / application: Modelling
    Keywords: Biostatistics, International networks, Scientific and operational challenges, Case studies
    Submitted at 28-Jul-2016 15:44 by Luc Bijnens
    Accepted (view paper)
    Today biostatisticians are confronted with a new kind of demand due to the massive collection of data by internal and external research laboratories. Internet connects scientists and laboratories who exchange big data. Over the last decades computing power increased overwhelmingly. For that reason classical solutions do not necessarily accommodate for model based drug development and time constraining interim analyses in the process of transforming data into knowledge. Statisticians play important roles in designing optimal experiments and help find the best statistical models to analyze the data coming from those experiments. Data are often integrated with other sources of information coming from observational data collection. Via publications of their work in peer reviewed journals biostatisticians can assist the regulators to create optimal and practical guidelines.
    For all those reasons pre-competitive collaborations in international networks amongst pharmaceutical, contract and academic statisticians are essential. Networks are modern operational solutions to contemporary challenges and opportunities in pharmaceutical research and development. Statistical innovation flourishes in networks because they are virtual, diverse and flexible and address todays’ statistical challenges. The aim of the talk is to show the tip of the iceberg of this multitude of scientific and operational challenges and opportunities. Case studies will be taken from discovery and early development research.