Curriculum Vitae: Leonard A Smith
Research Professor in Statistics
Director - Centre for the Analysis of Time Series
Professor of Statistics
Director, Centre for the Analysis of Time Series
Department of Statistics
London School of Economics
London WC2A 2AE
Senior Research Fellow (Mathematics),
Pembroke College, Oxford OX1 1DW
Employment
October 2004-present: Professor of Statistics (Research), London School of Economics (LSE)
May 1992-present: Senior Research Fellow (mathematics), Pembroke College, and Research Associate, Mathematics Institute, University of Oxford, Oxford, England
October 2001-present: Director of the Centre for the Analysis of Time Series (CATS), LSE
March 2000-September 2004: Reader (Research), Department of Statistics, LSE
May 1996-July 1996: Visiting Professor, Interdisziplinares Zentrum fur Kognitive Studien, Universitat Potsdam, Potsdam, Germany
April 1990-May 1992: Post-doc, University of Warwick, England
November 1989-April 1990: Visiting Scientist, Laboratoire de Meteorologie Dynamique du CNRS, Ecole Normale Superieure, Paris, France
June 1987-November 1989: Joint Post-doc between Department of Applied Mathematics and Theoretical Physics and Department of Pure Mathematics and Mathematical Statistics. Cambridge University. Cambridge, England
Education
PhD Columbia University, Physics, November 1987
Thesis title: Lacunarity and Chaos in Nature. Advisor: Prof. E. A Spiegel
MPhil Physics, Columbia University, May 1983
MA Physics, Columbia University, May 1982
BS (with High Honors) Physics, Mathematics & Computer Science, University of Florida, June 1980
^
Research Interests
The development of a coherent framework for using imperfect models to better understand and predict real dynamical systems; this includes model improvement and the use of model outputs. Experimental design of forecast studies using large complex models and their interpretation and application for policy support and decision making (e.g. climateprediction.net, DEMETER and ENSEMBLES). A major focus in the next few years is to extend such a framework in the context of Indistinguishable States (see my joint work with Kevin Judd) which has a firm mathematical basis, an arguable philosophical case and is deployable in practice. Our approach has been applied both for relatively simple physical systems and for operational weather models. Indistinguishable states provides a context for contrasting the dynamics of (1) physical systems, (2) analytic approximations of physical systems (equations of motion), and (3) the numerical approximations of these equations (including the dynamics of numerics). Diagnostic nonlinear time series analysis, including determining the data requirements of, and remaining uncertainty in, estimated scaling exponents; the dynamics of uncertainty in perfect models; a coherent definition of noise; the comparative evaluation of competing modelling paradigms; ultimate limits to predictability; the dynamics of, stability of, and fraud detection in industrial gambling systems; and a variety of applications to laboratory, medical, industrial, economic and natural systems ranging from the Earth's climate to the UK electricity grid.
^
Professional Activities & Honours
2005 SAMSI University Fellow (Data Assimilation); Statistical and Applied Mathematical Sciences Institute, Research Triangle, NC. www.samsi.info 2003 Fitzroy Prize from the Royal Meteorological Society 2002 Selby Fellowship from the Australian Academy of Science World Meteorological Organization, Expert Team on Verification (Member) THORPEX, International Science Steering Committee (Co-Chair for Societal and Economic Impacts, 2001-2004) (see www.mmm.ucar.edu/uswrp/programs/thorpex.html) climateprediction.net Scientific Steering Committee Smith Institute for Industrial Mathematics and System Engineering (www.smithinst.ac.uk) Scientific Advisory Panel. Oct 2002 - present Secretary of the European Geophysical Society (hereafter, EGS): Section on Nonlinear Processes. May 1994 - May 2004 Editor: Nonlinear Processes in Geophysics. May 1993 - 1997; special issue guest editor in 2001 and 2003 Scientific Editor and Contributor: The Earth and Physical Sciences Encyclopedia, Marshall Cavendish Corp. New York. A reference text for American high-school students Fellowship: Woods Hole Oceanographic Institution Summer School in Geophysical Fluid Dynamics. Summer 1984 Evaluation of research proposals for Australian, British, Canadian, Dutch, German and American research agencies
^
Selected Invited Lectures (not including seminars or contributed papers)
'20 years of Nonlinear Dynamics in Geosciences', Rhodes, June 2006. "Weather Forecasting: it's about Dynamics, it's not about Statistics". (AMS & EGU) American Meteorological Society, Ed Lorenz Symposium, San Diego, January 2005. "Insight of the Second Kind: A 21st Century View of Ed Lorenz's Visions of Predictability from the 1950's Onward" UK Met Office, Exeter, October 2004. "Seeing Through Ensembles", the opening talk of the week- long WMO Meeting on Ensemble Methods EU Policy Group, Seville, June 2004. "A Rational Role for Blind Fear in the Face of Scientific Risk Assessment" American Mathematical Society, Baltimore, January 2003. "Indistinguishable States: An Unified Approach to the Analysis and Prediction of Dynamical Systems Given Imperfect Nonlinear Models" 2002 European Centre for Medium Range Weather Forecasting (ECMWF) Seminar on Predictability, Reading, September 2002. "End-to-End Ensemble Enhanced Economic Estimation" The 2002 Selby Lectures: Public Lectures in seven cities across Australia, June-August 2002. "Science as Uncertainty: from Fractals in Forecasting, to Chaos in Climate Change" ( www.maths.uq.edu.au/rcw/Selby) Dynamics Days 2001, Dresden, June 2001. "Uncertainty from a time series" Sackler Colloquium (National Academy of Sciences), Los Angeles, March 2001. "Lifting the Excuse of Chaos: Predictability, Uncertainty and Error" The Sustainables (hosted by Royal Dutch Shell), Den Hague, November 2000. "Climate Change: The State of the Art(s) of Modelling" Symposium 2000 (hosted by IEEE), Lake Louise, Canada, October 2000. "Limits to Predictability in 2000 and 2100" Philosophy in Chaos? Kellogg College, Oxford, December 1999. A series of four lectures over two days, given jointly with Peter Smith Centre de Physique des Houches, France: 1998 Winter School "Mathematical and Physical Tools for Climate Dynamics" (3 Lectures) January 1998 Enrico Fermi School of Physics: 1996 Summer School at Varrenna, Italy. "Nonlinearity: Chaotic, Non-chaotic, and Random Dynamics." (3 Lectures) July 1996 British Ecological Society Workshop on "The analysis of non-linear dynamics in short ecological time series": University of Leicester, England. (3 Lectures) March 1996 XXI European Geophysical Society General Assembly: The Hague. "On Fleeting Shadows, Optimal Ensembles and Model Realism" May 1996 1995 European Centre for Medium Range Weather Forecasting (ECMWF) Seminar on Predictability: Reading, UK. "Accountability and Error in Ensemble Prediction of Baroclinic Flows." September 1995 Fields Institute Workshop. Nonlinear Dynamics and Time Series: Building a Bridge Between the Natural and Statistical Sciences. Montreal, Canada. "Ensemble Forecasting of Chaotic Systems." July 1995 6th International Meeting on Statistical Climatology: Galway, Ireland. "Ensemble Predictions and Chaotic Systems." June 1995 RSS 94: Royal Statistical Society International Conference, Newcastle. "Linear Intuitions in Nonlinear Dynamics: Missed Information from Least-squares solutions." Sept. 1994 San Diego '94: Society of Photo-Optical Instrumentation Engineers Annual Meeting, San Diego. "Predicting Predictability." July 1994 Royal Society Meeting: Chaos and Forecasting, London. "Nonlinear Prediction and Optimal Forecasting (The utility of knowing where you're at)." March 1994 Detecting Chaos in Real Data: CIBA Foundation Meeting, London. "Extrinsic limits on dimension calculations (What do we want to know?)." March 1994 American Statistical Association: 153rd Annual Meeting, San Francisco. "The Significance of Nonlinear Dynamical Systems Analysis: In the Lab and Averaged over the Earth." August 1993 Emerging Techniques in Signal Processing: Institute for Mathematics and its Applications, Warwick. "Nonlinear Signal Processing." December 1992 British Association Science Festival '92: BAAS, Southampton. "Extracting Predictions from Chaos." August 1992
Recent Consultancies
British Energy (also when trading as Nuclear Electric); DeutscherWetterdienst, European Centre for Medium-range Weather Forecasts, EDF Energy, IG Index, Royal Dutch Shell, SAIC, Scottish & Southern Energy, Scottish Power, The Smith Institute, UK Passport Agency.
Public Understanding of Science
I have done about half a dozen radio interviews in the UK and Australia, including one program of the Material World (Predicting the Unpredictable, aired April 11, 2002, see http://www.bbc.co.uk/radio4/science/thematerialworld_20020411.shtml
I have been actively involved with the climateprediction.net project since its beginnings, and am working with The Weather Channel with the goal of including uncertainty information in their weather forecasts. I have worked with the BBC on (and appear in the pilot of) the proposed weekly program "Turnedoutnicelyagain", which also has the aim of introducing the public to the advantages (and limits) of ensemble weather forecasting. Our work has been featured in New Scientist (cover story) and other popular academic magazines.
I have a long standing interest in the public understanding of science. Between 1978 and 1984, I taught a number a special courses each consisting of 6 to 10 classes, covering either astronomy, mathematics or computer science for gifted students in the primary grades (ages 6 to 12). These were given in various Public Schools in Gainesville, Florida and the New Lincoln School in New York City. Since that time I have given a number of lectures to school children and the general public, the latter at the invitation of the British Society for the Advancement of Science and Oxford's Department of Continuing Education, and the Australian Academy of Science, amongst others.
Further information
Further information is available on the Centre for the Analysis of Time Series (CATS) website
^
|
