Document Sample

C:\Docstoc\Working\pdf\bcca3c54-731a-4baa-865d-bb062508f3f7.xls Last printed on: 7/22/2011 Page 1 / 26 A B C D E F G H The numbers refer to the Form fields on the ISI 1 Proposal Form i. ISI Committee Affiliation: name of the ISI Committee ii. ISI or Section Affiliation: ISI, IAOS, IASC, IASS, IASE and/or Bernoulli Society iii. proposals to be Considered by the General Topics Committee Back to the Proposal Form 2 9.Short description of the 10.Name(s) of the 5.Date of 6.ISI/ISI Section 8.Title of Proposed 1.Name 3.E-mail address 7.Name of the society Invited Paper Meeting topic proposed submission representation Invited Paper Meeting incl references organiser(s) 3 application of distributed lag model to estimation 4 Atif Ali Yassin Ali atifyassin2@yahoo.com 9/23/2004 iii consumption function More: See the concepts, and Generally for the derivation of Q5 orgumentents on atomic energy, radius and velocity of an models,and the theory based electron are caliculated by on the circular motion of equlising the centrifugal and General proposition, ISI, 5 G.V. Sekhar 11/11/2004 i , iii atoms coulombic attraction forces. Indian ISI/IASS/IAOS member More: See Q6 Statistical offices have more and Eric Schulte Nordholt / more information and researchers Dutch / Senior Remote access and remote want this information for their statistician at Statistics 6 Eric Schulte Nordholt ESLE@CBS.NL 11/18/2004 ii execution analyses. Netherlands Karl Pearson can be characterized as the father of statistics. 2007 is his 150th Ida H. Stamhuis, Vrije birthyear. In 2004 a biography of Universiteit, The Karl Pearson written by Theodore Netherlands, Vice 7 Ida H. Stamhuis stamhuis@few.vu.nl 12/3/2004 iii Karl Pearson's 150th birthday M. Porter has been published. President IAOS The purpose of this invited paper session is to bring together researchers whose recent theoretical and applied work is of interest to both survival analysis and econometrics. These two areas share Dorota M. Dabrowska many non- and semiparametric regression Non-and semiparametric models, as well as special observational from University of regression models in survival schemes accouting for missing data and California, Los Angeles 8 Dorota M. Dabrowska dorota@brahms.ph.ucla.edu 12/7/2004 ii Bernoulli Society analysis and econometrics biased sampling problems. (nationality - USA) Last updated on: February 15, 2005 C:\Docstoc\Working\pdf\bcca3c54-731a-4baa-865d-bb062508f3f7.xls Last printed on: 7/22/2011 Page 2 / 26 A I J K L M N O P Q R The numbers refer to the Form fields on the ISI 1 Proposal Form i. ISI Committee Affiliation: name of the ISI Committee ii. ISI or Section Affiliation: ISI, IAOS, IASC, IASS, IASE and/or Bernoulli Society iii. proposals to be Considered by the General Topics Committee (name) 12.Proposed under the programme of: 2 More 13 Possible S C E S I SI lli So t 11.Names of possible c. es IAO IAS IAS IAS ou Collaborating Gu 1.Name speakers and their 3.E-mail address 7.Name of the society rn Societies Be nationalities and affiliations: 3 See G5 4 Atif Ali Yassin Ali x x x More: But according to Loney's, Zemansy's theory the caliculation is done in slight converse. They said that in circular motion See Q5 centrifugal force is not equal to coulombs force. Since the cetrifugal force is pseudo force. It consisted only to the mass of the electron.and we neglect the motion of nucles. But actually nucles has some osscilation. It is negleted only when the mass is negligable or infinetily. But the mass of the electron considered to be 1/1836 of the photon. So we must consider the mass of the nucles. So, we take the mass of the electron as reduced mass mM/(M+m),to consider the mass 5 G.V. Sekhar G.Veera Sekhar, Indian x of the nuclues. And about the energy states there are more concepts to discuss like this. See G6 To protect the ISI/IASS/IAOS member confidentiality of the respondents statistical disclosure control techniques More: Seedeveloped. have been Q6 Traditionally only tables had to be protected. Additionally some statistical offices have delivered (partly) protected microdata. At some institutes also the possibility exists to work on-site on microdata. With the new remote techniques Speakers of e.g. Statistics the data stay at the premises of the statistical institutes and the researchers stay at their institutes. Protected data could Denmark, Australian Bureau of be analysed with remote access (direct access to statistical data without a check) and other data could be analysed with Statistics and many other remote execution (only safe results of the analyses will be sent back). The organisational and technical aspects of the 6 Eric Schulte Nordholt organisations x x x different remote options is an interesting topic for an IP session in Lisbon in 2007. Theodore M. Porter, Professor of History, University of California in Los Angelos; Stephen M. Stigler, Professor of Statistics, University of Chicago, and President ISI; Eileen Magnello, Wellcome Institute London; Ida H. Stamhuis, Vrije Universiteit 7 Ida H. Stamhuis Amsterdam and Vice President IAOS x More: See Q8 See i8 Four researchers accepted invitation. Gang Li from University of California, Los Angeles (Chinese), Michael Kohler from Stuttgart University (nationality -German), Gang Li (UCLA) will present work on partial linear regression with censored data and applications to analysis of health Joel Horowitz from Northwestern insurance data. Michael Kohler (Stuttgart University) will present work on nonparametric regression estimation from University, Chicago (nationality - censored data and applications to prediction. USA) and Jens Perch Nielsen Joel Horowitz (Northwestern University) will discuss estimation in nonparametric additive models with known and from Codan in Copenhagen unknown link function. Jens Perch Nielsen (Codan) will talk about estimation of extremes from filtered data with a view to 8 Dorota M. Dabrowska (nationality -Danish) x fire insurance in London commercial market. Last updated on: February 15, 2005 C:\Docstoc\Working\pdf\bcca3c54-731a-4baa-865d-bb062508f3f7.xls Last printed on: 7/22/2011 Page 3 / 26 A B C D E F G H 9.Short description of the 10.Name(s) of the 5.Date of 6.ISI/ISI Section 8.Title of Proposed 1.Name 3.E-mail address 7.Name of the society Invited Paper Meeting topic proposed submission representation Invited Paper Meeting incl references organiser(s) 3 The topic of this section consists More: See in statistical methods for Q9 regression models under shape restrictions. This includes estimation (i.e. parametric nonparametric estimation under monotonicity, convexity or other constraints) and testing (i.e. testing for monotonicity or other shape restrictions) in such models. Particular emphasiz is put methods for (nonparametric) regression models with 9 Holger Dette holger.dette@rub.de 12/14/2004 ii Bernoulli Society Shape-Restricted Regression multivariate predictors. Holger Dette, Germany Longitudinal studies have received intensive attention More: See recently. Many efforts have been Q10 devoted to statistical modelling Clearly, to avoid the wrong conclusions due to the use of improper models, model checking Lixing Zhu, Chinese, Model Checking and for the established models is The University of Hong 10 Lixing Zhu lzhu@hku.hk 12/15/2004 ii Bernoulli Society Longitudinal Studies important. Kong There has recently been a large More: See published literature on new Q11 objective Bayesian methods (where, as in frequentist statistics, the result only depends on the assumed model and the Jose M. Bernardo, 11 Jose M. Bernardo jose.m.bernardo@uv.es 12/16/2004 ii, iii ISI, Bernoulli Society Objective Bayesian Statistics observed data). Spain, ISI More: See Q12 Semi-supervised learning refers to classification problems in which the training data contains both labeled as well as unlabeled observations. Such problems arise naturally in many contexts, Bin Yu (China) such as handwriting recognition University of California, 12 Bin Yu binyu@stat.berkeley.edu 1/4/2005 ii, iii Bernoulli Society Semi-Supervised Learning and document classification. Berkeley More: See Q13 We are concerned by multiple table analysis and the exploratory side of the analysis. This implies Exploratory multiple table that we are going to deliberatedly A Morin french IRISA 13 Annie Morin amorin@irisa.fr 1/4/2005 ii ISI analysis avoids the PLS methods. Rennes Last updated on: February 15, 2005 C:\Docstoc\Working\pdf\bcca3c54-731a-4baa-865d-bb062508f3f7.xls Last printed on: 7/22/2011 Page 4 / 26 A I J K L M N O P Q R More 13 Possible S C E S I SI lli So t 11.Names of possible c. es IAO IAS IAS IAS ou Collaborating Gu 1.Name speakers and their 3.E-mail address 7.Name of the society rn Societies Be nationalities and affiliations: 3 See G9 More: See Q9 References: Ghosal, Subhashis; Sen, Arusharka; van der Vaart, Aad W. Testing monotonicity of regression. (English) Ann. Stat. 28, No.4, 1054-1082 (2000) Hall, Peter; Huang, Li-Shan Nonparametric kernel regression subject to monotonicity constraints. (English) Ann. Stat. 29, No.3, 624-647 (2001) J. Wellner, P. Hall, V. Spokoiny, L. Hall, Peter; Müller, Hans-Georg Duembgen, E. Mammen, L. Order-preserving nonparametric regression, with applications to conditional distribution and quantile function estimation. 9 Holger Dette Davies, A. Van der Vaart x (English) J. Am. Stat. Assoc. 98, No.463, 598-608 (2003). See G10 More: See Q10 In this session, the newly developed approaches of goodness-fit-testing will be introduced, which are used to test whether some covariate effects in the model follow certain parametric forms, or more generally, to check the adequacy of a parametric model. References: 1).Crainiceanu, C. M. and Ruppert, D. (2004). Restricted likelihood ratio tests in Dr. Colin Wu, Dr. C. M. nonparametric longitudinal models, Statistica Sinica, 713-730. 2).Xu, W, L. and Zhu, L. X.(2004). Goodness-of-fit Tests 10 Lixing Zhu Crainiceanu, and Dr. W. L. Xu x for a Varying-Coefficients Model in Longitudinal Studies. Submitted for publication. See G11 This includes the development of a general theory (Berger, Bernardo and Sun, 2004) which permits the definition of More: See a reference prior from first principles, the analysis of the behaviour under repeated sampling of objective Bayesian Q11 procedures (Datta and Sweeting, 2004), the development of new powerful methods in model selection (Bayarri, 2004; James O. Berger, USA, ISI; Pericchi, 2004) model criticism (Bernardo and Rueda, 2002) and applications with complex, often hierarchical models in Trevor Sweeting, UK; Susie many different fields (eg climatology, image analysis, metrology, bioinformatics or reliability). An invited paper session Bayarri, Spain; Raul Rueda, focusing on Objective bayesian methods would give delegates an opportunity to get acquainted with the latest 11 Jose M. Bernardo Mexico x x developments in this emerging topic. More: See Q12 See G12 Mikhail Belkin (University of Chicago), David Mease (University of California, Berkeley), Peter In most cases, unlabeled observations are relatively easy to obtain, while the labeled observations are quite costly and Bickel (University of California, consequently relatively scarce. The main objective in such problems is to learn classification rules using information Berkeley), Neil Lawrence from the unlabeled observations in some way to provide lower misclassification error rates than rules learned using only (Sheffield University), Yaacov the labeled documents. Some researchers have had success using ideas from spectral graph theory. Other approaches Ritov (The Hebrew University of include EM-based methods and transductive support vector machines. This session will focus on new practical and 12 Bin Yu Jerusalem) x theoretical advances for problems in semi-supervised learning. More: See See G13 Q13 We want to focus on the use of factorial analysis and clustering methods for multiple table. It is not a new area but the applications are increasing in sensorial analysis, marketing and so on and the exploratory methods are still there with some new developments even if many users focus on Path modelling. H Kiers Netherlands , J Pagès References : Escofier B, Pagès J multiple factor analysis in computational statistics and data analysis in 1994 France, A Lazraq Marocco, V Caroll J D A generalization of canonical correlation analysis 13 Annie Morin Vinzi Italy x to three or more sets of variables 1968 Last updated on: February 15, 2005 C:\Docstoc\Working\pdf\bcca3c54-731a-4baa-865d-bb062508f3f7.xls Last printed on: 7/22/2011 Page 5 / 26 A B C D E F G H 9.Short description of the 10.Name(s) of the 5.Date of 6.ISI/ISI Section 8.Title of Proposed 1.Name 3.E-mail address 7.Name of the society Invited Paper Meeting topic proposed submission representation Invited Paper Meeting incl references organiser(s) 3 More: See Q14 Once, a politician declared : ―I don‘t need statistics; I only need indicators‖ ! Among some categories of users, the opinion is generally received that simple and synthetic indicators are more useful than complex and multi- dimensional statistics that are Do users need indicators or supposed to be designed for 14 Jean-Louis Bodin jean-louis.bodin@adetef.finances.gouv.fr 1/7/2005 ii IAOS, IASS statistics ? researchers. Jean-Louis Bodin More: See Q15 Over the last decade wavelets have revolutionized nonparametric regression, especially for non-homogeneous functions and begun to make a large impact for locally stationary time series. However, it is not always clear how to apply wavelet (and Fourier) methods to irregularly spaced data especially Guy Nason, British, 15 Guy Nason g.p.nason@bristol.ac.uk 1/10/2005 ii Bernoulli Society Lifting in Statistics in higher dimensions. University of Bristol, UK More: See Q16 Over the last three decades there is a growing interest in stochastic analysis due to its broad applications on financial market. The main focus of this Ching-Tang Wu section is on stochastic analysis (Taiwan; National Stochastic Analysis and and its application on financial University of Kaohsiung, 16 Ching-Tang Wu ctwu@nuk.edu.tw 1/11/2005 ii Bernoulli Society Financial Mathematics mathematics. Kaohsiung, Taiwan) Recent SPRUCE workshop which was held in Lisbon, March 2004, More: See identified current issues in space Q17 time modelling (http://spruce.deio.fc.ul.pt ; reports of the working groups) as being: 1) Change of support (Data collected at different levels of aggregation) 2) Misaligned and incompatible data 3) Data Measured or simulated at Kamil Feridun Turkman Spatial/Temporal Models and different space-time scales University of Lisbon, 17 Kamil Feridun Turkman kfturkman@fc.ul.pt 1/11/2005 ii, iii ISI Methods ... Portugal Last updated on: February 15, 2005 C:\Docstoc\Working\pdf\bcca3c54-731a-4baa-865d-bb062508f3f7.xls Last printed on: 7/22/2011 Page 6 / 26 A I J K L M N O P Q R More 13 Possible S C E S I SI lli So t 11.Names of possible c. es IAO IAS IAS IAS ou Collaborating Gu 1.Name speakers and their 3.E-mail address 7.Name of the society rn Societies Be nationalities and affiliations: 3 More: See G14 See There will be ideally three papers : Q14 - one promoter of a well-known Statisticians are used to calculate synthetic indicators for many years (GDP, or CPI are synthetic indicators) but there is synthetic indicators: the Indicator a growing tendency to introduce more and more indicators, not only quantitative but also qualitative. of Human Development (Dr. Amatya Sen or Ms Sakiko Fukuda- The use of synthetic indicators may raise many questions : Parr) - what are the characteristics of a ―good‖ synthetic indicator (in particular qualitative indicators) ? - one politician aware of these - is their apparent simplicity an illusion or reality ? problems (Michel Rocard, Member - is benchmarking, based on a comprehensive system of several key indicators, may be effective for policy purposes of the European Parliament) (see the Palermo Forum on ―Statistics, Knowledge and Policy‖ recently organised by OECD) - one staff member of the - indicators are pushing users – in particular the media - to make rankings : are such rankings really useful ? are they statistics department of the dangerous ? European Central Bank or of the - indicators may lead to ―statistically driven decisions‖ (e.g. Maastricht indicators), which confuses the relations and the statistics department of OECD. debate between decision-makers and statisticians and paves the way to political pressures on statisticians - etc. The discussants being persons The proposed session will try to better define the respective role of synthetic indicators and more complex statistics with responsible of National Statistical the idea of demonstrating that users (businesses, governments, non-profit organisations, advocacy groups, citizens) 14 Jean-Louis Bodin Councils or Heads of NSOs. x x need both tools to make informed decisions. More: See See G15 Q15 Bernard Silverman, British, University of Oxford, UK; Rainer Lifting is a new (late '90s) generalization of wavelets that allows multiscale decompositions of irregularly spaced data von Sachs, German, Universitie (also in higher dimensions) that is generating great excitement in the statistical multiscale community. Lifting has great Catholique de Louvain, Belgium; potential for irregularly spaced data for regression and time series analysis but also for spatial data and the multiscale Guy Nason, British, University of analysis of networks. Bristol, UK; Maarten Jansen, Belgian, T.U. Eindhoven, A Lisbon 2007 session would possibly be one of the 1st at a major conference. This would permit interaction between Netherlands; Veronique Delouille, experts but also publicising the benefits of lifting to a wider audience. Belgian, Royal Observatory of 15 Guy Nason Belgium. x This is a very new area with new methods developing but with wide potential applicability. See G16 More: See Professor Dr. Hans Foellmer Q16 (German; Humboldt Universitaet zu Berlin, Berlin, Germany), The most important scholars in this subject will publish their new findings. Their works are expected to shed lights on risk Professor Dr. Marc Yor (France; management and other related financial and insurance issues. Universite Pierre et Marie Curie, Paris, France), Professor Dr. The key references include: H. Foellmer and A. Schied: "Robust Preferences and Convex Risk Measures", M. Atlan, H. Walter Schachermayer (Austria; Geman, D. B. Madan, and M. Yor: "Correlation and the pricing of risks", W. Schachermayer and F. Delbaen: "The University of Technology Vienna, Fundamental Theorem of Asset Pricing for Unbounded Stochastic Processes", and W. Schachermayer: 16 Ching-Tang Wu Vienna, Austria) x "Optimal Investment in Incomplete Financial Markets". See G17 More: See Q17 Christopher Wikle, University of ... Missouri, USA; Doug Nychka 4)Extremes of Random fields National Center for Atmospheric 5)Non-separable space-time covariance models Research, USA; Jef Teugels, 6)Principles for selecting the appropriate levels of aggregation Katholieke Universiteit Leuven, 7) Non-Gaussian data Belgium; Jim Zidek, University of 8) Useful metrics for comparing different models and assessing adequacy of models British Columbia, Canada; Noel 9) Computationally efficient implementation of space-time models Cressie, Ohio State University, USA; Patrick Brown , University of This reseach field is very active and no doubth all these issues will be pursued durig the coming years. It is therefore Lancaster, UK; Valerie Isham, usefull and desirable to include a session in the 56th ISI session in Lisbon to assess the advances made in this area. 17 Kamil Feridun Turkman University College, UK. x Last updated on: February 15, 2005 C:\Docstoc\Working\pdf\bcca3c54-731a-4baa-865d-bb062508f3f7.xls Last printed on: 7/22/2011 Page 7 / 26 A B C D E F G H 9.Short description of the 10.Name(s) of the 5.Date of 6.ISI/ISI Section 8.Title of Proposed 1.Name 3.E-mail address 7.Name of the society Invited Paper Meeting topic proposed submission representation Invited Paper Meeting incl references organiser(s) 3 Adequate bias reduction in the More: See semi-parametric estimation of Q18 rare events‘ parameters provides usually smoother paths, smaller mse at optimal levels and a smaller sensitivity to changes in location. One then needs to estimate properly the bias function, or the second order parameters in that bias. The Bias Reduction in the adequate accommodation of the Jan Beirlant, Katholieke Estimation of Parameters of bias of Hill‘s estimator has been University Leuven, 18 M. Ivette Gomes ivette.gomes@fc.ul.pt 1/11/2005 ii Rare Events extensively addressed. Belgium Spatial Extrmes is the next challenge for research in extreme value theory, now that the finite- dimensional case is well understood. The theoretical foundations have been laid and a number of models have been developed. But the area still has to grow to maturity. Applied data are waiting for proper analysis Thomas Mikosch, (rainfall, pollution data). The main university of 19 Laurens de Haan ldhaan@few.eur.nl 1/11/2005 ii Bernoulli Spatial Extremes argument is "new area". Copenhagen, Denmark More: See Ray Hilborn - School of Q20 Exploitation of wild animal Aquatic and Fishery Sciences, University of populations requires quantitative Washington, Box 357980, scientific advice, to prevent stock Seattle, WA 98195-79800, depletion. Preservation of natural USA. habitats and of high levels of rayh@u.washington.edu; biodiversity is a concomitant key Nancy Lo - NOAA issue. Recently, it has been Southwest Fisheries recognized the importance of Science Center, P.O. Box providing measures of uncertainty 271, Lajolla, California in outputs from assessment 92038 nlo@ucsd. edu; Ransom A. Myers - models and also of including in Department of Biological them the possibility of changes in Sciences, Life Sciences environment. Uncertainty has Centre 1355 Oxford Street been incorporated mainly through Dalhousie University Recent Developments in sensitivity analysis, bootstrap, Halifax, Nova Scotia, Management of Animal bayesian methods and spatio- Canada, B3H 4J1 20 M. Lucilia Carvalho mlcarvalho@fc.ul.pt 1/11/2005 ii ISI or Bernoulli Populations temporal analysis. Ransom.Myers@dal.ca The application of design of experiment methods to large computer experiments is a new chalange emerging from physical simulation. First, there is no measurement errors and second the size of the designs is usually very large (more than 10^5 units and more than 50 variables). This session will try to Design for large computer give an overview of the state of Azais J.M. and Gamboa 21 Azais J.M and Gamboa F. gamboa@cict.fr 1/11/2005 ii Bernoulli experiments art in the area. F. Last updated on: February 15, 2005 C:\Docstoc\Working\pdf\bcca3c54-731a-4baa-865d-bb062508f3f7.xls Last printed on: 7/22/2011 Page 8 / 26 A I J K L M N O P Q R More 13 Possible S C E S I SI lli So t 11.Names of possible c. es IAO IAS IAS IAS ou Collaborating Gu 1.Name speakers and their 3.E-mail address 7.Name of the society rn Societies Be nationalities and affiliations: 3 See G18 More: See Beirlant, Jan — Katholieke Q18 Universiteit Leuven, Belgium; Feuerverger, Andrei — University of Toronto, Canada; Gomes, M. Ivette — University of Lisbon, Portugal; Hall, Peter — Canberra University, Australia; Martins, M. The external estimation of the shape second order parameter enables us keeping under adequate control the asymptotic João — Technical University of variance of the tail index reduced bias‘ estimators. Only quite recently attention has been paid to the scale second order Lisbon, Portugal; Neves, M. parameter. The external estimation of such a parameter enables us to estimate, in a bias reduced way, any rare events‘ Manuela — Technical University parameter, without increasing the optimal asymptotic variance. This is a way to overpass the old trade-off between of Lisbon, Portugal; Peng, Liang variance and bias. Further work is expected in the next years, together with applications in the fields of insurance and 18 M. Ivette Gomes — Atlanta University, USA. x finance. J.Einmahl, university of Tilburg, The Netherlands; T.T. Pereira, University of Lisbon, Portugal; L. de Haan, Erasmus university, The Netherlands; J. Tawn, university of 19 Laurens de Haan Lancaster, UK. x More: See See G20 M. K. McAllister - Renewable Q20 Resources Assessment Group, Department of Environmental science and Technology, Imperial College, Royal School of mines Building, Prince Consort Road, London, SW7 2BP, U.K; A. Punt - School of Aquatic and Fishery Sciences, University of Washington, Box 355020, Seattle, WA 98195-5020, USA. aepunt@u.washington.edu; H. Caswell. - Biology Department, Woods Hole Oceanographic As a result models have been growing in complexity and in basic internal assumptions, often not clearly stated and Institution, Woods Hole, MA without explicit relationships with the biology and dynamics of the resources. Those highly complex models become then 02543-1049, USA. less transparent and difficult to access by the users. We intend to promote a debate on the trade off between the use of 20 M. Lucilia Carvalho hcaswell@whoi.edu x x such highly structured models as key tools for management or as a test tool to evaluate robustness of simpler models. Santner T. (Ohio state University), Owen A .B (Stanford University), 21 Azais J.M and Gamboa F. Dette H (Bochum University), x Last updated on: February 15, 2005 C:\Docstoc\Working\pdf\bcca3c54-731a-4baa-865d-bb062508f3f7.xls Last printed on: 7/22/2011 Page 9 / 26 A B C D E F G H 9.Short description of the 10.Name(s) of the 5.Date of 6.ISI/ISI Section 8.Title of Proposed 1.Name 3.E-mail address 7.Name of the society Invited Paper Meeting topic proposed submission representation Invited Paper Meeting incl references organiser(s) 3 In statistics, the term "official More: See data" denotes data collected in Q22 censuses and statistical surveys by National Statistics Institutes, as well as administrative and registration records collected by government departments and local authorities. They are used to Donato Malerba, produce "official statistics" for the Dipartimento di purpose of making policy Informatica, University of 22 Paula Brito mpbrito@fep.up.pt 1/12/2005 ii IAOS or ISI Mining Oficial Data decisions. Bari, Italy More: See The need to extend standard Q23 exploratory, statistical and graphical data analysis methods to more complex data, has been increasing. This is the case of data concerning more or less homogeneous classes or groups of individuals - second-order objects or macro-data - instead of single individuals - first-order objects or micro-data. Symbolic Data Analysis allows the extension of classical data Paula Brito, Faculdade de analysis techniques to the Economia, Universidade do 23 Paula Brito mpbrito@fep.up.pt 1/12/2005 ii IASC Symbolic Data Analysis analysis of second-order objects. Porto, PORTUGAL More: See Q24 In Statistics of Extremes it is always assumed the validity of a first order condition. i.e., it is assumed that the model underlying data is in the domain M. Isabel Fraga Alves, of attraction of an Extreme Value DEIO, Faculdade de distribution. And, whenever we Ciências, Universidade de intend to accommodate bias of Lisboa, Portugal (or estimators, we need to go further, Laurens de Haan, assuming second order Erasmus University of Testing Extreme Value conditions, specifying the rate of Rotterdam, The 24 M. Isabel Fraga Alves isabel.alves@fc.ul.pt 1/12/2005 ii ISI or Bernoulli Conditions convergence in the first order. Netherlands) Last updated on: February 15, 2005 C:\Docstoc\Working\pdf\bcca3c54-731a-4baa-865d-bb062508f3f7.xls Last printed on: 7/22/2011 Page 10 / 26 A I J K L M N O P Q R More 13 Possible S C E S I SI lli So t 11.Names of possible c. es IAO IAS IAS IAS ou Collaborating Gu 1.Name speakers and their 3.E-mail address 7.Name of the society rn Societies Be nationalities and affiliations: 3 See G22 More: See Q22 David J. Hand, Imperial College, London, UK; Michael May, Fraunhofer Institute for Very few applications of data mining techniques in the discovery of new models or patterns in official data sets have Autonomous Intelligent Systems, been reported. Although the application of data mining techniques to official data has great potential both in the public Sankt Augustin, Germany; Carlos and in the private sectors, mining official data requires a challenging methodological research in AGGREGATED DATA, Marcelo, National Institute of GEO-REFERENCIATION, DATA QUALITY, TIMELINESS, METADATA. This is only a sample of the most challenging 22 Paula Brito Statistics (INE), Lisbon, Portugal x x problems. The Invited Paper Meeting aims to draw added attention to this field and stimulate further research. See G23 More: See Q23 Edwin Diday, University of Paris- IX Dauphine , Paris, France; Lynne Billard, University of Georgia, Athens, USA; Carlo Lauro, Università degli Studi di Symbolic data extend the classical tabular model, allowing multiple, possibly weighted, values for each descriptive Napoli "Federico II", Naples , Italy; attribute. Symbolic data occur when summarising huge sets of data or describing the underlying concepts - a town, a Francisco de Carvalho, Federal scenario of accidents - of a database. Symbolic Data Analysis provides useful tools to analyse aggregated data in the University of Pernambuco (UFPE), field of official statistics. Symbolic Data Analysis underwent great improvement with the European projects ―SODAS‖ and 23 Paula Brito Recife, Brazil x x ―ASSO‖ and the development of a software package. More: See Q24 See G24 De Haan, Laurens — Erasmus University of Rotterdam, Holland; Falk, M. — Wurzburg University, Germany; Fraga Alves, M. Isabel — Lisbon University, Portugal; Hasofer, A. M. — Victoria University of Technology; Husler, Jurg — Bern University, Switzerland; Mahron, F. — Testing the first order condition has been performed a few years ago, under a parametric framework, and methodology Eichstat University, Germany; for the separation of three regions, negative, null and positive tail index, has then been advanced. More recently, this Teugels, Jef; Katholieke trilemma has been partially solved under a semi-parametric framework. A lot of work on the topic has been recently Universiteit Leuven, Belgium; under development and the portuguese research team in this area even thinks sensible to revisit, now under a semi- Wang, J. Z. — University of parametric framework, most of the topics developed in a parametric set-up. This will for sure lead to further important 24 M. Isabel Fraga Alves Western Sydney, Australia x x developments in the next few years. Last updated on: February 15, 2005 C:\Docstoc\Working\pdf\bcca3c54-731a-4baa-865d-bb062508f3f7.xls Last printed on: 7/22/2011 Page 11 / 26 A B C D E F G H 9.Short description of the 10.Name(s) of the 5.Date of 6.ISI/ISI Section 8.Title of Proposed 1.Name 3.E-mail address 7.Name of the society Invited Paper Meeting topic proposed submission representation Invited Paper Meeting incl references organiser(s) 3 Teaching statistics, we use real- world problems and data. 1) Using random number tables, More: See each student may build his own Q25 sample from the real dataset and gets his own estimation of a statistic of interest. We merges all the results using a spreadsheet. Using a macro- instruction, each student is able to simulate a number of random samples and to calculate the statistics value of each sample; Teaching inferential statistics he can study the distribution, and demography with calculate confident intervals, test Jean-Hugues CHAUCHAT, simulations involving real- different hypothesis, draw the French, Université Lumière- 25 Jean-Hugues Chauchat Jean-Hugues.Chauchat@univ-lyon2.fr 1/12/2005 ii IASE world data and problems power curve, and so on. Lyon2 Much of statistical inference defies exact analysis, and relies More: See on asymptotics. The usual first order asymptotics based primarily Q26 on the central limit theorems and laws of large numbers allow many competing candidates which are all first order efficient. Higher order asymptotics enables one to distinguish between all these candidates. Such asymptotics is relevant both in frequentist and Bayesian frameworks, and has many Malay Ghosh; American; 26 Malay Ghosh ghoshm@stat.ufl.edu 1/13/2005 ii Bernoulli Society Higher Order Asymptotics potential applications. University of Florida Spatial and Spatio-Temporal Statistics have become the most important methodologies for a large diversity of disciplines (Climatology, Environmental Science…). The most important feature that must be taken into account when working with spatial data is the dependence structure that observed data may exhibit. Scientists from other areas are not always aware of this fact and spatially dependent data have been thought as independent. This situation no longer subsists and Spatial Statistics are nowadays playing a leading role. Apart from the practical implications, many efforts have been devoted to the development of different techniques for estimation and prediction with spatial data. Estimation of the variogram, non- stationary processes analysis, goodness-of- fit testing and resampling methods for spatial data have become research topics Wenceslao González- of great interest. The latest advances and Manteiga. Full professor. applications of these methodologies in Department of Statistics. Recent Advances in Spatial different fields would be described and University of Santiago de 27 Dr. Wenceslao González-Manteiga wences@zmat.usc.es 1/14/2005 ii Bernoulli Society Statistics discussed in this meeting. Compostela. Last updated on: February 15, 2005 C:\Docstoc\Working\pdf\bcca3c54-731a-4baa-865d-bb062508f3f7.xls Last printed on: 7/22/2011 Page 12 / 26 A I J K L M N O P Q R More 13 Possible S C E S I SI lli So t 11.Names of possible c. es IAO IAS IAS IAS ou Collaborating Gu 1.Name 3.E-mail address speakers and their 7.Name of the society rn Societies Be nationalities and affiliations: 3 See G25 More: See Q25 (1) Doing things by him-self from a real population, and not only to use computers to simulate a theoretical distribution, J-H CHAUCHAT, French, students understand each concepts of inference. Univ.Lyon2; M BOURDEAU, 2) In order to understand utility of statistics on social life, it is useful to simulate demographic forecasting for different real Canadian, Univ.Montréal; E countries, using real mortality and fecundity tables for each country. Such notion as active/inactive ratio and its evolution 25 Jean-Hugues Chauchat NEUWIRTH, Austria, Univ.Vienna x in different countries became easily understandable. See G26 More: See Q26 Gauri Sankar Datta; American, Univ. of Georgia; Rahul Mukerjee, Indian, Indian Institute of The proposed session covers both frequentist and Bayesian higher order asymptotics, and includes some of the current Management; Trevor Sweeting, active areas of research such as small area estimation, empirical likelihood and inference based on partial prior 26 Malay Ghosh British, University College, London. x information. See i27 More: See NOEL CRESSIE. Affiliation: Q27 Distinguished professor of Mathematical and Physical Sciences. Department of Statistics. The Ohio State University. Nationality: Australian citizen with permanent immigrant status in the USA. Research topics: theory and applications of spatial and spatio-temporal stochastic modeling; Bayes and empirical Bayes methods. MONTSERRAT FUENTES.Affiliation: Associate professor.Department of Statistics. North Carolina State University. MICHAEL STEIN. Affiliation: Nationality: Spanish citizen with permanent Immigrant status in the USA.Research topics: Spatial-temporal models, Director of CISES (Center for spatial statistics, spectral methods, computational methods for large data sets, applications in climate and air quality Integrating Statistical and modeling, oceanography, atmospheric and environmental science. Environmental Science) and full PETER J. DIGGLE. Affiliation: Professor.Department of Mathematics and Statistics. Lancaster University. Nationality: professor at the Department of British citizen.Research topics: Point processes. Most of the research is motivated by applications in the biomedical or Statistics.Department of Statistics. environmental. University of Chicago. Nationality: SOUMENDRA N. LAHIRI. Affiliation: Professor.Department of Mathematics and Statistics. Iowa State University. USA.Research topics: statistical Nationality: Indian citizen. (No information about immigrant status)Research topics: Environmental Statistics. models and methods for spatial Resampling 27 Dr. Wenceslao González-Manteiga and spatial-temporal processes. x methods for dependent data. Spatial Statistics. Last updated on: February 15, 2005 C:\Docstoc\Working\pdf\bcca3c54-731a-4baa-865d-bb062508f3f7.xls Last printed on: 7/22/2011 Page 13 / 26 A B C D E F G H 9.Short description of the 10.Name(s) of the 5.Date of 6.ISI/ISI Section 8.Title of Proposed 1.Name 3.E-mail address 7.Name of the society Invited Paper Meeting topic proposed submission representation Invited Paper Meeting incl references organiser(s) 3 In multivariate stat and econometrics, boundary More: See estimation (support of a Q28 multivariate pdf) is important. In production analysis, the frontier is the locus of optimal production plans. In finance, it corresponds to efficient assets combinations. Nonparametric estimators are based on the idea of enveloping at best the cloud of the Léopold SIMAR, multivariate observations (Data Belgian, Institute of Envelopment DEA, Free Disposal Statistics, University of 28 Léopold Simar simar@stat.ucl.ac.be 1/14/2005 ii Bernoulli Society Boundaries Estimation Hull FDH). Louvain-la-Neuve More: See This session is intended to Q29 present (1) new statistical models for describing two-loci genetic interaction in complex binary traits and their Bayesian analysis by MCMC methods; (2) new methods towards identification of differentially Carlos Daniel Paulino/ expressed genes in Cystic Portuguese/ Instituto Fibrosis(according an empirical Superior Técnico, Bayesian Methods in Bayes approach); Technical University of 29 Carlos Daniel Paulino dpaulino@math.ist.utl.pt 1/14/2005 ii, iii ISI, Bernoulli Society Biostatistics ... Lisbon Efficiency of a statistical test is a very useful notion for practitioners when they must select the best test among many competitors. For theoreticians its calculation poses many interesting challenging problems yanikit47@pisem.net OR connected with convex analysis, large deviations, variational calculus,etc. The aim of the session is to present and to Efficiency of Tests: Recent discuss the developments in the Yakov Nikitin, Russia, 30 Yakov Nikitin Yakov.Nikitin@pobox.spbu.ru 1/14/2005 ii Bernoulli Society Advances efficiency theory obtained in last years. St.Petersburg University The papers and possible a panel discussion could address some of the following points: A brief description of the available technologies; their impact on the courses; Eric Sowey (Australian); course as problem-driven rather than David Green (British); technique driven; the use of exploratory How have modern Neville hunt (British); data analysis; the choice of data; the time technologies changed the devoted to computer use; simulation for Adrian Bowman curriculum in introductory theorem verification; personnal work on (British); J.P. Florens 31 l Catherine Pardoux and Lucette Carterucette.carter@wanadoo.fr 1/14/2005 i: GdSF ii IASE; IASC courses real data analysis; problem of evaluation. (French). Information theory, a study on data compression and transmission, has its root in mathematical probability and statistics. It shares some fundamental ideas with statistical inference and it should be of interest to statisticians. Talks in this session will cover applications of information theory both in theoretical statistics (such as estimation and model selection) and in Information Theory and empirical statistical analysis (such as Feng Liang, ISDS, Duke 32 Feng Liang feng@stat.duke.edu 1/14/2005 ii Bernoulli Society Statistics network tomography). University Last updated on: February 15, 2005 C:\Docstoc\Working\pdf\bcca3c54-731a-4baa-865d-bb062508f3f7.xls Last printed on: 7/22/2011 Page 14 / 26 A I J K L M N O P Q R More 13 Possible S C E S I SI lli So t 11.Names of possible c. es IAO IAS IAS IAS ou Collaborating Gu 1.Name 3.E-mail address speakers and their 7.Name of the society rn Societies Be nationalities and affiliations: 3 See G28 More: See Q28 Alois Kneip, German, Univeristy of The statistical properties of these estimators are difficult to handle and the bootstrap is one way to produce practical Mainz, A. Tsybakov, Russian, inference. New nonparametric estimators with better statistical performances have been proposed. Hall-Simar (2002) University of Paris VI, Paul Estimating a Changepoint, Boundary or Frontier in the presence of noise, JASA. Kneip-Park-Simar (2003) Asymptotics Wilson, US, University of Texas, for DEA estimators in nonparam front models. Aragon-Daouia-Thomas-Agnan (2005), Nonparametric Front Estim: a A. Daouia, Marocco, University of conditional quantile approach, Econometric Theory. Jeong-Simar (2005), Linearly interpolated FDH estimator for Toulouse, S-O Jeong, Korean, nonconvex frontiers 28 Léopold Simar University of Seoul X See G29 More: See Q29 Nuno Sepúlveda/ Portuguese/ Instituto Gulbenkian de Ciência; Lisete Sousa/ Portuguese/ University of Lisbon; Giovani Silva/ Portuguese/ IST, Technical ... University of Lisbon; Paulo (3) a new semiparametric regression analysis of marked point processes in survival settings; Soares/ Portuguese/ IST, (4) new Bayesian methods of log-linear analysis for informatively incomplete categorical data with prior information 29 Carlos Daniel Paulino Technical University of Lisbon X X about the original parameters of the Multinomial model for the observable counts. T.Ledwina,T.Inglot (Poland,Wroclaw), W.Kallenberg (The Netherlands, Twente), M.Arcones (USA, Binghampton), Q.-M.Shao(USA, Oregon), M.Hallin (Belgium, Brussels), N.Stepanova (Canada, Carleton), A.Mogulski (Russia, Novosibirsk), 30 Yakov Nikitin etc. X X 31 Catherine Pardoux and Lucette Carter X X X Yuhong Yang, University of Minnesota; Gang Liang, University of California, Irvine; Peter Grunwald, CWI, Netherlands; 32 Feng Liang Feng Liang, Duke University; X Last updated on: February 15, 2005 C:\Docstoc\Working\pdf\bcca3c54-731a-4baa-865d-bb062508f3f7.xls Last printed on: 7/22/2011 Page 15 / 26 A B C D E F G H 9.Short description of the 10.Name(s) of the 5.Date of 6.ISI/ISI Section 8.Title of Proposed 1.Name 3.E-mail address 7.Name of the society Invited Paper Meeting topic proposed submission representation Invited Paper Meeting incl references organiser(s) 3 In recent years, the quality of data available in finance has improved More: See substantially, especially with the Q33 advent of high freqency data (in some cases observations as often as several times per second). This provides new opportunities for using sophisitcated statistcal approaches to such data. The three proposed speakers are leaders in this effort, a revolution that is in the process of Per Mykland (Norway transforming transforming USA), University of 33 Per Mykland mykland@galton.uchicago.edu 1/14/2005 ii Bernoulli Society Statistics and Finance financial econometrics. Chicago Interacting particle systems is now an important area of Probability Theory. This session will present a few topics of new researchs. In particular, the relation of these systems and quasi stationary distributions, with queing theory and with fase transition in 34 Pablo A. Ferrari pablo@ime.usp.br 1/14/2005 ii Interacting particle systems Statistical Mechanics. Pablo A. Ferrari More: See Q35 Data Mining, Quality-Control Mining, and Text Mining make use of all the 'traditional statistical procedures' plus the newer Artificial Intelligence / Machine Learning Algorithms. Although a very small segment of the business-world has used this technology, to its clear advantage, for 10 years, most of the business, governmental,and industrial world has only started using "data mining technology" during the past four years, and academia is only beginning DATA MININER, QC-Mining, to discover the 'distinct advantages' Gary Miner, Ph.D.; Text Mining and Comparison Linda A. Miner, Ph.D.; Gary D. Miner, Ph.D. and of AI / Machine Learning and enhanced predictive power of these technologies, utilizing various Paolo F. Ricci, Ph.D. 35 Linda A. Winters-Miner, Ph.D. gary@statsoft.com 1/14/2005 iii Statistical Algorithms areas of Bayesian statistics. LLM The speakers of this session will present research in which robustness will be a basic philosophical approach to statistics, motivated by the realities of data analysis. Both theoretical and applied research will be presented, and the interaction between Luisa T. Fernholz statistical theory and statistical computing will be highlighted as a way to create new (Temple University, methodologies that offer powerful tools for USA, 36 Graciela Boente gboente@dm.uba.ar 1/14/2005 ii,iii Bernoulli Society Robustness and Data Analysis analyzing data. fernholz@temple.edu) Soft indicators of public sector More: See performance (such as satisfaction and Q37 trust, and as opposed to "hard" indicators such as budget and output) more and more influence the debate on reforms of government (Bouckaert & Van de Walle, 2003). In order to measure changes of satisfaction and trust, governments increasingly apply surveys. In this contribution, we point at a fundamental weakness of soft indicators, which is their sensitivity to the mode of data collection (face-to-face, mail) and to the context of The Impact of Survey surveying (e.g., government initiative Methodology and Context on versus university initiative; long versus 37 Dr. Jarl K. Kampen jarl.kampen@soc.kuleuven.ac.be 1/15/2005 iii Soft Indicators of Government short survey). Last updated on: February 15, 2005 C:\Docstoc\Working\pdf\bcca3c54-731a-4baa-865d-bb062508f3f7.xls Last printed on: 7/22/2011 Page 16 / 26 A I J K L M N O P Q R More 13 Possible S C E S I SI lli So t 11.Names of possible c. es IAO IAS IAS IAS ou Collaborating Gu 1.Name 3.E-mail address speakers and their 7.Name of the society rn Societies Be nationalities and affiliations: 3 See G33 More: See Q33 Technicques that have been invented by the proposed speakers draw heavily on both finance, nonparametric statistics, and insights from, Jianqing Fan (USA, Princeton), probability, such as stochastic calculus. Topics that have been pursued Lan Zhang (China, Carnegie by the speakers include the combination of localization in time and Mellon University), Michael space, the insight that securities prices have ``measurement error'', 33 Per Mykland Sorensen (Denmark) X and the theory of likelihood inference and estimating equations. Errico Presutti (Roma Tor Vergata), Stefano Olla (Paris Dauphine), Amine Asselah (marseille), James Martin (CNRS), Serguei Popov (Sao Paulo), 34 Pablo A. Ferrari Servet Martinez (Chile) X More: See See G35 Q35 Gary Miner and Linda Miner - USA; Paolo Ricci, Italy, USA & Gary D. Miner, Ph.D. and Australia; Sachin Lahoti, India; 35 Linda A. Winters-Miner, Ph.D. Kiron Mathew, India X Thus, it is important that the statistical world becomes fully congnizant of the usefulness of these technologies. 1. David Tyler, Rutgers University, USA 2. J. Cabrera (Spain) Rutgers 36 Graciela Boente University X X See G37 More: See Q37 Statistical analyses reveal that the mode effect could amount upto a 10% change of item nonresponse and a 9% change of the level of trust, whereas the context effect could lead upto a 19% change of the bivariate associations of trust. The observed changes of both measures of central tendency and of associations affect conclusions of survey research in 37 Dr. Jarl K. Kampen X isolated studies. Last updated on: February 15, 2005 C:\Docstoc\Working\pdf\bcca3c54-731a-4baa-865d-bb062508f3f7.xls Last printed on: 7/22/2011 Page 17 / 26 A B C D E F G H 9.Short description of the 10.Name(s) of the 5.Date of 6.ISI/ISI Section 8.Title of Proposed 1.Name 3.E-mail address 7.Name of the society Invited Paper Meeting topic proposed submission representation Invited Paper Meeting incl references organiser(s) 3 Seasonal adjustment is still a More: See challenging topic in particular for Q38 official statisticians and remains an active field for research. For example, 5 sessions were dedicated to seasonal adjustment during the last JSM in Toronto (August 2004). Every year, a lot of contributions are made in order to improve existing methods and to propose New Developements in new solutions to seasonal Dominique Ladiray, 38 Ladiray Dominique dominique.ladiray@insee.fr 1/19/2005 ii Seasonal Adjustment Methods adjustment problems. French, INSEE More: See Assumptions on shapes and Q39 constraints for multivariate inference can be viewed as an alternative to dimension reduction, or complexity reduction. Many fields fit into this framework. An example is functional data analysis, where it makes sense to align functions according to their shapes. Other work has qualitative features as target of the analysis. They can be seen as low dimensional objects, or functionals of the distribution, Shapes and constraints in resulting in an implicit complexity 39 Wolfgang Polonik wpolonik@ucdavis.edu 1/27/2005 ii Bernoulli Society statistical inference reduction. More: See Q40 See A39 Assumptions on shapes and constraints for multivariate inference can to a certain extend be viewed as an alternative to dimension reduction, or maybe better complexity reduction. This of course if of great importance in modern statistical inference. There are many fields in statistics which fit into this framework. An example is functional data analysis, where it makes sense to "align" functions according to their shapes. Another direction of work aims at qualitative features of the underlying distributions as target of the analysis. These features can be interpreted as "low dimensional objects", or functionals of the Shapes and constraints in distribution, and hence there is 40 Wolfgang Polonik ii Bernoulli Society statistical inference an implicit complexity reduction. Last updated on: February 15, 2005 C:\Docstoc\Working\pdf\bcca3c54-731a-4baa-865d-bb062508f3f7.xls Last printed on: 7/22/2011 Page 18 / 26 A I J K L M N O P Q R More 13 Possible S C E S I SI lli So t 11.Names of possible c. es IAO IAS IAS IAS ou Collaborating Gu 1.Name 3.E-mail address speakers and their 7.Name of the society rn Societies Be nationalities and affiliations: 3 See G38 More: See Q38 For just a few examples: 1 - The US Bureau of Census will soon release the new version of X-12-ARIMA which incorporates both X-12 and SEATS, David Findley (US Bureau of 2 - The OECD Task Force on Data Presentation and Seasonal Adjustment came up in 2004 with recommendations for a Census), Agustin Maravall (Bank metadata and reporting manual, of Spain), Benoît Quenneville 3 - New methods to seasonally adjust daily, weekly and other periodic data are under development at the INSEE, (Statistics Canada), Dominique I already had contacts with A. Maravall (Bank of Spain), D. Findley (US Bureau of Census) and B. Quenneville (Statistics 38 Ladiray Dominique Ladiray (INSEE, France) X Canada) who support this initiative and are ready to contribute to this session. See G39 More: See Q39 G. Walther, Stanford University; H.-G. Mueller, UC Davis, P. Hall, One such methodology is bump hunting. A particular example is the PRIM algorithm of Friedman and Fisher. For ANU,;A. B. Tsybakov, University univariate cases see G. Walter, L. Davies and A. Kovac. Paris VI; L. Davies, University of Another shape aspect in PRIM is that the algorithm is working with rectangles which in itself constitutes a shape Essen, Germany; Ming-Yen constraint. Other work using shape restrictions is related to classification, for a classifyer for a two class problem can be 39 Wolfgang Polonik Cheng, National Taiwan University X interpreted as a set (Tsybakov, 2004). See G40 More: See Q40 See A39 One particular instance of such an methodology is bump hunting, and an particular example for such a method is the PRIM algorithm of Friedman and Fisher. There is another shape aspect in PRIM, namely, the algorithm is working with rectangular shapes which in itself constitutes a shape constraint or complexity reduction. Other work using shape G. Walther, Stanford University restrictions in this sense are related to classification, since a classifyer for a two class problem can be interpreted as a P. Hall, ANU set (see Tsybakov, 2004). A. B. Tsybakov, University Paris VI L. Davies, University of Essen All the above are problems of multivariate nature. However, even in the univariate case there are interesting open Ming-Yen Cheng, National Taiwan problems in the area of bump hunting, as can be seen in recent work by G. Walter or L. Davies and A. Kovac. It is work 40 Wolfgang Polonik University X along these lines that I envision to be presented in this section. Last updated on: February 15, 2005 C:\Docstoc\Working\pdf\bcca3c54-731a-4baa-865d-bb062508f3f7.xls Last printed on: 7/22/2011 Page 19 / 26 A B C D E F G H 9.Short description of the 10.Name(s) of the 5.Date of 6.ISI/ISI Section 8.Title of Proposed 1.Name 3.E-mail address 7.Name of the society Invited Paper Meeting topic proposed submission representation Invited Paper Meeting incl references organiser(s) 3 More: See Q41 Organizer: Prof. Michelle Liou, Institute of Statistical Science; The use of modern techniques in Chair: Prof. John Aston, brain science makes greater Institute of Statistical demands on statistical Science methodologies, now more than ever. Research into brain Discussants: Prof. structure/ connectivity and neural Fernando Lopes da circuits has been conducted using Silva, Swammerdam techniques such as magnetic Institute for Life resonance imaging (MRI), Sciences; positron emission tomography Prof. Philip E. Cheng, (PET), and electro/ Institute of Statistical magnetoencephalography Science, Academia 41 Michelle Liou mliou@stat.sinica.edu.tw ii Bernoulli Society (EEG/MEG). Sinica More: See Q42 The analysis of microarray data is the focus of much of recent statistical research. However, each microarray experiment is commonly viewed in isolation. It is generally believed that for a more thorough understanding of complex biological systems we Rebecka Jornsten need to pool knowledge from (Swedish), Department multiple microarray data sets as of Statistics, Rutgers 42 Rebecka Jornsten rebecka@stat.rutgers.edu ii Bernoulli Society well as other data sources. University, USA There is a long list of applications of information theoretic entropic- More: See optimization principles in Q43 physical, biological, social, managemen, business and engineering sciences and technology.The reason for their versality is that all systems are information theoretic and any principle that helps us in 1. Dr. Om Parkash, measuring minimum and G.N.D. University, maximum information are bound Amritsar, India. to be useful.The proposed 2. Dr. Parmil Kumar, Entropy Optimization invitedpaper meeting will give Jammu University, Principles with Applications in new insights into application of Jammu 43 Prof. D.S.Hooda ds_hooda@rediffmail.com iii N.A. Statistics entropy optimisation principles. (J&K) This session will not only describe the statistics training programs in major Latin More: See American countries, but also Q44 emphasize the links (if any) among countries that share a common historical ancestry and language. The focus will be on Shrikant I. Bangdiwala, Training of Statisticians in relationships between Portugal USA, University of Iberoamerica and Iberia: with Brazil and between Spain North Carolina at What are the links and what and Spanish-speaking Latin Chapel Hill - ISI 44 Shrikant I. Bangdiwala kant@unc.edu 2/13/2005 ii IASE are the gaps? American countries. member, IASE member Last updated on: February 15, 2005 C:\Docstoc\Working\pdf\bcca3c54-731a-4baa-865d-bb062508f3f7.xls Last printed on: 7/22/2011 Page 20 / 26 A I J K L M N O P Q R More 13 Possible S C E S I SI lli So t 11.Names of possible c. es IAO IAS IAS IAS ou Collaborating Gu 1.Name speakers and their 3.E-mail address 7.Name of the society rn Societies Be nationalities and affiliations: 3 More: See Q41 See G41 Prof. Keith J Worsley, Department of Mathematics and Statistics, McGill University, ―Random fields, Statistics must be able to identify scientific patterns inherent in noisily spatio-temporal data sets collected using modern astrophysics, brain imaging and imaging techniques This session will be concerned with recent topics and challenges in both statistical theory and ‗bubbles‘‖ methods for understanding brain structure and its dynamic systems. Prof. Tohru Ozaki, The Institute of References: Statistical Mathematics, ―Spatial Martinex-Montes, E, Valdes-Sosa, PA, Miwakeichi, F, Goldman, RI & Cohen, M (2004). Concurrent EEG/fMRI analysis temporal modeling of EEG/fMRI by multiway partial least squares, NeuroImage, 22, 1023-1034. for the study of brain connectivity‖ Worsley, KJ, Taylor, JE & Tomaiuolo, F (2004). Unified univariate and multivariate random field theory. NeuroImage, in Prof. Pedro A. Valdes-Sosa, press. Cuban Neuroscience Center, ―Concurrent EEG/fMRI analysis‖ Yamashita, O, Galka, A, Ozaki, T, Biscay, R & Valdes-Sosa, PA (2004). Recursive least squares solution for dynamic 41 Michelle Liou X inverse problems of EEG generation, Human Brain Mapping, 21, 221-235. We list some important biological questions of interest here. See G42 1. Is gene co-expression conserved among organisms? More: See Attempts have been made at answering these questions by looking at the relation between pairwise correlations between Q42 genes in one organism and the pairwise correlations between their corresponding orthologoues in other species. This type of analysis may improve the functional annotation of genes. 2. Are the regulatory regions of functional groups of genes conserved among organisms? This questions relates to investigating gene expression data and DNA sequence data jointly. This type of analysis may also shed light on phylogeny. Preliminary analysis suggests that phylogenetic distances cannot be inferred from expression data alone. 3.Cross-species extrapolation. Many genomic studies focus on human diseases. However, human studies often have serious limitations. Can reliable and relevant information be extracted from large-scale animal model studies? To answer these types of questions, microarray data sets from multiple species have been analyzed. The experimental conditions, and set of genes in each study, have only partial overlap, which makes such cross-species analysis particularly challenging. (to be expanded) Sunduz Keles, Department of Our goal for this session is to present a rich set of research problems related to the analysis of multiple platforms- Statistics, University of Wisconsin, multiple species data. The data formats present special challenges and much input is needed from the statistical USA community to develop better analysis methods. Saeed Tavazoie, Department of We propose to invite speakers from both statistics and biology, to present initial work and their view of the problems at Molecular Biology, Princeton hand. 42 Rebecka Jornsten University, USA X See G43 More: See Q43 Prof. Aman Ullah, University of California. 2.Prof. Lynne M.Batten, Deeakin University, Australia. REFERENCES 3. Dr. Omparkash, 1. D.V.Gokhale and S.Kullback: The information in contingency tables. Macel-Decker, New York, 1978. G.N.D.University, Amritsar. 2. D.S.Hooda and B.K.Hooda:On measurement of stochastic dependence in multivariate data.Indian Journal of Pure and 4. Dr. Parmil Kumar, Jammu Applied Mathematics, 32,2001, 801-16. University, Jammu (J&K). 3.J.N.Kapur and H.K.Keshavan: Generalized Entropy Optimization Principles with applications. Acdemic Press,New 5. Prof. B.K.Das, University of yorl,1992. 43 Prof. D.S.Hooda Delhi, Delhi (INDIA). X 4 S.Kullback: Information theory and statistics. John Wiley, New York,1959. ISI Committee 1. Raul Mentz, Universidad Nacional de Tucuman, Argentina; 2. Julio Singer, Universidade de More: See Sao Paulo, Brazil; 3. Sergio Q44 Muñoz, Universidad de La Frontera, Chile; 4. Victor Perez Major themes to address: Abreu, Instituto Nacional de 1. Why there are not many statisticians in Latin America and how to change it. See G44 Estadistica, Mexico; 5. Ma. Nelcy 2. Is the training too technical/theoretical, less applied? Rodriguez, Universidad Nacional, 3. Who enters the field of statistics in such countries? Colombia; 6. Carmen Batanero, As ISI seeks to expand its membership in countries where there has been small representation, this topic is of special Universidad de Granada, Spain; 7. interest since the meeting is in Portugal IASE Organizing 44 Shrikant I. Bangdiwala President - Portuguese Statistical X Speakers will be invited from Portugal and Brazil; and Spain and Argentina, Chile, Mexico, Colombia, and Costa Rica Committee Last updated on: February 15, 2005 C:\Docstoc\Working\pdf\bcca3c54-731a-4baa-865d-bb062508f3f7.xls Last printed on: 7/22/2011 Page 21 / 26 A B C D E F G H 9.Short description of the 10.Name(s) of the 5.Date of 6.ISI/ISI Section 8.Title of Proposed 1.Name 3.E-mail address 7.Name of the society Invited Paper Meeting topic proposed submission representation Invited Paper Meeting incl references organiser(s) 3 More: See This session aims to use the Q45 widespread interest in sport to demonstrate variability phenomena that arise in many fields of study. Instead of demonstrating statistical tools with sport applications, the session shows how, starting with a realistic sport situation or practical issue, investigation with simple tools can reveal surprising results, advancing students intuitive understanding of 45 Larry Weldon weldon@sfu.ca 2/13/2005 ii IASE randomness. Curriculum in undergraduate More: See Statistics is often affected by Q46 advances in research and success in new fields of application. This topic explores ways in which the development of technology affects the curriculum, either by eliminating some topics, or adding topics, or moving topics or concepts earlier in the curriculum. A trivial example is the nearly complete elimination of Robert Gould, USA, the "calculator" formula for UCLA, Department of 46 Robert Gould rgould@stat.ucla.edu 2/13/2005 ii IASE Technology and Curriculum sample standard deviation. Abstract Statistics In the emerging field of statistical More: See education research, more Q47 attention is being paid to how students reason about data as an aggregate group, coming to understand a distribution in a global sense, rather than focusing on individual values. Students need a firm understanding of the concept of distribution in order to interpret graphical representations of data, to compare groups of data, recognize covariation, to think statistically about the role of Research on Reasoning variation of data, and making Joan Garfield, US, 47 Joan Garfield jbg@umn.edu 2/14/2005 ii IASE about Distribution inferences about data. University of Minnesota Economists are faced with the More: See task of absorbing a constant flow Q48 of new knowledge in their field of practice and conduct their own research. 2.Economical students, postgraduate trainees must possess the tools and abilities to understand the issues related to designing a valid scientific study and to comprehend the quantitative aspects of economical research. They must also acquire the ability to interpret and critically evaluate this research in order to Ruslan Motoryn, determine how and to what extent Ukraine, Prof.,dr. of Statistical Training and the findings can be applied to Kiev National University 48 Ruslan Motoryn motoryn@i-c.com.ua 2/14/2005 ii IASE Education of Economists their own activities. of Economics Last updated on: February 15, 2005 C:\Docstoc\Working\pdf\bcca3c54-731a-4baa-865d-bb062508f3f7.xls Last printed on: 7/22/2011 Page 22 / 26 A I J K L M N O P Q R More 13 Possible S C E S I SI lli So t 11.Names of possible c. es IAO IAS IAS IAS ou Collaborating Gu 1.Name 3.E-mail address speakers and their 7.Name of the society rn Societies Be nationalities and affiliations: 3 See G45 More: See Q45 Some relevant references of the proposed speakers: 1. de Silva, Basil M. , and Swartz, Tim B. (1997), ``Winning the coin toss and the home team advantage in one-day international cricket matches'', New Zealand Statistician [Became: @J(AusNZJSt)], 32 (2) , 16-22 2. Clarke, Stephen R. , and Norman, John M. (1995), ``Home ground advantage of individual clubs in English soccer'', The Statistician, 44 , 509-521 3. Lock, Robin (2002), ``Offense and defense as predictors of team success'', Stats. The Magazine for Students of 45 Larry Weldon Statistics, 0 (34) , 22-25 More: See Q46 See G46 Mark Hansen, US, UCLA; Arthur A more controversial proposal (IASE Roundtable 2004) is the elimination of boxplots from middle-school curriculum. Bakker, Netherlands, Utrecht; Tim This topic will also welcome speculation about how the curriculum might be changed in the future to adapt to new 46 Robert Gould Erickson, US, Fathom Software X technologies to handling massive data and very fast computations. More: See See G47 Q47 Maxine Pfannkuch, NZ, University of Auckland; Dani Ben-Zvi, Israel, University of Haifa; Andee Rubin, USA, TERC; Chris Reading, AU, University of New England; Beth The papers in the suggested session present current research on different aspects of reasoning about distribution. They Chance, USA, Cal Poly San Luis examine students at different age levels as well as secondary teachers' conceptions. The discussion will include uses of Obispo; Dave Pratt, UK, University technological tools to promote distributional reasoning, assessment of this reasoning, and curricular implications at all 47 Joan Garfield of Warwick X educational levels. unknown at this time More: See G48 Q48 Members of ISI: Prof. Yury Ivanov (Russia); Prof. Guido Ferrari (Italy);Prof. Helmut Maier 3.This session offers views and experiences of leading economic statistical educators from various backgrounds who will (Germany);Prof. Aleksandr Zelias report on different approaches and styles of teaching economic statistics. (Poland);Prof. Byung-Soo Kim 4.Relations between statistics and economics need to address statisticians as academics, teachers and professionals. 48 Ruslan Motoryn (Korea). X 5.References: Economic statistics(2005)Textbook,Kyiv. unknown at this time Last updated on: February 15, 2005 C:\Docstoc\Working\pdf\bcca3c54-731a-4baa-865d-bb062508f3f7.xls Last printed on: 7/22/2011 Page 23 / 26 A B C D E F G H 9.Short description of the 10.Name(s) of the 5.Date of 6.ISI/ISI Section 8.Title of Proposed 1.Name 3.E-mail address 7.Name of the society Invited Paper Meeting topic proposed submission representation Invited Paper Meeting incl references organiser(s) 3 Many teachers of statistics believe that simulations help students to understand fundamental concepts of probability and statistical More: See inference. Q49 Empirical educational research concerning this claim is rare. The session should bring together empirical researchers and initiate in advance a collaborative review of research that can be presented in the session to a wider public. Also, specific new reserach approaches, namely qualitative studies beside quantitative ones, Rolf Biehler, University Research on the use of should be included and discussed of Kassel, Dept of simulation in teaching with regard to their scope and Mathematics / 49 Rolf Biehler biehler@mathematik.uni-kassel.de 2/14/2005 ii IASE statistics and probability limitations. Informatics In an ever evolving technological world, statistics education is also More: See changing. The Internet alone Q50 IASE provides a rich source of resources for teaching statistics and innovative teaching Name of the Guest approaches. This session highlights recent projects which Society: Consortium assist statistics educators in for the Advancement facing the challenge of determining the best approaches of Undergraduate for utilizing the diversity of Optimizing Internet-based teaching resources and methods Statistics Education Resources for Teaching available through using the 50 Ginger Holmes Rowell rowell@mtsu.edu 2/14/2004 ii (CAUSE) Statistics Internet. Ginger Holmes Rowell More: See Q51 In observational data, association is certainly not a clear sign of causation. Yet most students studying statistics are in majors such as business or the social Observational Studies, sciences where observationally Milo Schield, USA, ASA Confounding and Multivariate based associations are used as Section on Statistical 51 Milo Schield milo@pro-ns.net 2/14/2004 ii IASE Thinking evidence for causal connections. Education More: See Q52 Some New Methods for Obtaining Baseline National Data in Developing Countries and in Some Countries of Europe, Mainly Survey Methods for Obtaining Concerning Agricultural, Alvaro González 52 Alvaro González Villalobos alvarun@fibertel.com.ar 2/14/2005 ii IASS Baseline National Data Population and Housing Data Villalobos Last updated on: February 15, 2005 C:\Docstoc\Working\pdf\bcca3c54-731a-4baa-865d-bb062508f3f7.xls Last printed on: 7/22/2011 Page 24 / 26 A I J K L M N O P Q R More 13 Possible S C E S I SI lli So t 11.Names of possible c. es IAO IAS IAS IAS ou Collaborating Gu 1.Name 3.E-mail address speakers and their 7.Name of the society rn Societies Be nationalities and affiliations: 3 See G49 More: See Q49 The session should also reflect on the various computer programs that are available for simulation (from JavaApplets to tool software), identify most promising uses of simulation, and specify educational conditions under which simulation seems to have the most benefits. Joachim Engel (Hannover, Also, specific new reserach approaches, namely qualitative studies beside quantitative ones, should be included and Germany), Carmen Batanero discussed with regard to their scope and limitations. (Granda, Spain), Cliff Konold The session should also reflect on the various computer programs that are available for simulation (from JavaApplets to (Univeristy of Massachusetts at tool software), identify most promising uses of simulation, and specify educational conditions under which simulation 49 Rolf Biehler Amherst, USA) X seems to have the most benefits. IASC is a possibility More: See See G50 Q50 One of these projects is CAUSEweb, a digital library of undergraduate statistics education resources, which is being Neville Davies, Great Britain, developed and operated by the Consortium for the Advancement of Undergraduate Statistics Education (CAUSE) Chris Wild, New Zealand, Jane through funding from the US National Science Foundation. Presenters will describe issues related to providing teaching Watson, Australia, Jura Puranen, resources via the Internet such as keeping resources current with rapidly changing technologies, facilitating learning with I do not know of any 50 Ginger Holmes Rowell Finland X X a web-based environment, and training teachers to use available technology. at this time. See G51 More: See Q51 Unless they take a course in multivariate regression, students may never realize that an association can increase, Howard Wainer (US), James decrease or even reverse (Simpson‘s paradox) after taking into account the influence of a confounder – or that a Nicholson (Ireland) & Jim Ridgway statistically significant association can become statistically insignificant. Should students be taught these topics in the (UK), George Cobb (US), John introductory course? Can these topics be taught without including a whole course in multivariate regression? Can Brignell (UK), Steve Raudenbush students be considered statistically literate if they are unaware of these matters? Is multivariate thinking a key (US), Dick Frazer (US), Peter component of statistical literacy? The speakers will comment on these questions and on the importance of confounding 51 Milo Schield Holmes (UK). X in statistical education. See G52 More: See Q52 References: Encuestas Agrícolas y Encuestas a Hogares: estrategias comunes (Sara Novaro), FAO Interamerican Meeting, Panama, 2002 Elisabetta Carfagna (ISI/IASS), CPS Technical Paper 63-Appendix A (1997), "Maximizing PSU Overlap, http://www.bls.census.gov/cps Sara Novaro (IASS) and Mirane Ernst, L.R. (1995), "Maximizing and Minimizing Overlap of Ultimate Sampling Units". Bureau of Labor Statistics, 2 52 Alvaro González Villalobos M, Carrilho (IASS) X Massachusetts Ave, N.E, Room 3160, Washington DC 20212 IAOS Last updated on: February 15, 2005 C:\Docstoc\Working\pdf\bcca3c54-731a-4baa-865d-bb062508f3f7.xls Last printed on: 7/22/2011 Page 25 / 26 A B C D E F G H 9.Short description of the 10.Name(s) of the 5.Date of 6.ISI/ISI Section 8.Title of Proposed 1.Name 3.E-mail address 7.Name of the society Invited Paper Meeting topic proposed submission representation Invited Paper Meeting incl references organiser(s) 3 Statistical More: See Confidenciality, Q53 Research Purposes, Anonymisation Methodology Harala, R, and Reinikainen, A-L. (1994), ‗Statistical confidentiality and the What procedures to keep the use of safety of sensible records, either statistical data for from business or personal research purposes – registers. How far to go in the Finnish aspects‘, anonymisation and what Second International Statistical Confidentiality and methodologies to use in the Seminar on Statistical the Use of Statistical Data for preparation of the information to Confidentiality, 53 Pedro Corte Real pedro.real@ine.pt 2/11/2005 Research Purposes be used by researchers Luxembourg 1994. The aim of the session is to share experiences amongst organizations concerning best practises in quality reporting in the statistical process. Key words: quality components; quality reports; quality indicators; Maria João Zilhão Maria João Zilhão mjoao.zilhao@ine.pt Measuring Quality in statistical process; quality Margarida Madaleno, 54 Margarida Madaleno margarida.madaleno@ine.pt 2/10/2005 ii IAOS Statistical Processes assurance. both Portuguese The aim of the session/meeting is to share experiences amongst organizations concerning best practises in implementing systematic quality work: difficulties and resistances encountered; training; quality teams; general involvement of the employees; methods to overcome resistance; models used to implement systematic quality work (eg, ISO Norms; EFQM model; CAF, etc...). Maria João Zilhão Quality Management in Key words: Systematic Quality Maria João Zilhão 55 Margarida Madaleno mjoao.zilhao@ine.pt 2/10/2005 ii IAOS Official Statistics Work; TQM Margarida Madaleno Customer Relationship Management (CRM) The aim of the session is to share experiences amongst Organizations concerning best practises in Customer Relationship Management of statistical organizations, and instruments used to access customer satisfaction of the statistical products. Key words: Customer satisfaction Maria João Zilhão Customer Relationship surveys; Call centres; Customer Maria João Zilhão 56 Margarida Madaleno mjoao.zilhao@ine.pt 2/10/2005 ii IAOS Management (CRM) satisfaction Margarida Madaleno International Cooperation in Statistics The aim of the session is to share experiences concerning international cooperation activities, including provision of technical assistance to strengthen statistical capacity of other countries and organizations. Margarida Madaleno and Margarida Madaleno International Cooperation in Key words: international Conceição Veiga, both 57 Conceição Veiga margarida.madaleno@ine.pt 2/11/2005 ii IAOS Statistics cooperation; technical assistance Portuguese Last updated on: February 15, 2005 C:\Docstoc\Working\pdf\bcca3c54-731a-4baa-865d-bb062508f3f7.xls Last printed on: 7/22/2011 Page 26 / 26 A I J K L M N O P Q R More 13 Possible S C E S I SI lli So t 11.Names of possible c. es IAO IAS IAS IAS ou Collaborating Gu 1.Name 3.E-mail address speakers and their 7.Name of the society rn Societies Be nationalities and affiliations: 3 More: See See H53 Q53 Harala, R, and Reinikainen, A-L. (1996), ‗Confidentiality in the use of administrative data sources‘, Statistical Journal of the United Nations Economic Commission for Europe, 4/1996. Riitta Harala Statistics Finland (2004), Use of Registers and Administrative Data Sources for Statistical Purposes, Best Practices of Statistics Finland, Handbooks 45, Helsinki 53 Pedro Corte Real Anna-Leena Reinikainen 2004 Maria João Zilhão 54 Margarida Madaleno X Maria João Zilhão 55 Margarida Madaleno X Maria João Zilhão 56 Margarida Madaleno X Margarida Madaleno 57 Conceição Veiga X Last updated on: February 15, 2005

DOCUMENT INFO

Shared By:

Categories:

Tags:

Stats:

views: | 164 |

posted: | 7/22/2011 |

language: | English |

pages: | 26 |

Description:
Quantitative Models for Performance Evaluation and Benchmarking Data Envelopment document sample

OTHER DOCS BY psi16197

How are you planning on using Docstoc?
BUSINESS
PERSONAL

By registering with docstoc.com you agree to our
privacy policy and
terms of service, and to receive content and offer notifications.

Docstoc is the premier online destination to start and grow small businesses. It hosts the best quality and widest selection of professional documents (over 20 million) and resources including expert videos, articles and productivity tools to make every small business better.

Search or Browse for any specific document or resource you need for your business. Or explore our curated resources for Starting a Business, Growing a Business or for Professional Development.

Feel free to Contact Us with any questions you might have.