Contents

Homepage of the Institute:
http://www.rni.helsinki.fi
 

Electronic versions of this report:
http://www.rni.helsinki.fi/~llh/annual_report/2000
 
 
 
 
 
 

1  The Year in Review

Last year the research staff of the Institute was organized into three research divisions: Biometry, Mathematical Methods of Information Technology, and Mathematical Theory and Applications of Electromagnetic Fields. Professor Jukka Sarvas was on a sabbatical leave and Dr. Lasse Holmström acted as the Director of the Institute.

The scientific output of the institute and the number of completed advanced degrees both increased from the previous year. The total number of papers included in the annual report increased from 77 to 89 while the number papers in international refereed journals increased 28%, from 18 to 23 (cf. Figure 1). Doctoral degrees were completed by Mikko Sillanpää, Pasi Korhonen, and Andriy Andreev, all in Biometry. Simopekka Vänskä completed the degree of Licentiate in Philosophy in Applied Mathematics.

The total budget in 2000 was 6.9 mmk (million marks) of which the University of Helsinki provided 1.47 mmk, or 21%. This is up from the previous year's 18%, definitely a step in the right direction that reverses the 5-year declining trend of the University's share in our funding (see Figure 1). One should however point out that a sizable portion of University funding, 0.28 mmk, was granted only for special short-term projects that ended in 2000. The University's share without these project funds was 1.19 mmk or 17% of our total budget. The Institute is therefore funded mostly by outside sources. The largest portions (Figure 2) of outside funding came from Academy of Finland research grants (25%) and the Graduate Schools funded by the Ministry of Education (24%). During the whole or parts of the past year the Institute held 13 positions in 6 different Graduate Schools. This testifies for success in researcher training and also reflects the highly interdisciplinary nature of our research work. Research contracts with industrial partners and other sources, most importantly the National Public Health Institute, both accounted for 15% of the total budget.

The Institute has taken part in two recent scientific evaluations. The first one was the Research Assessment Exercise of the University of Helsinki in 1999 and the second one was the Academy of Finland Evaluation of Mathematics carried out last year. We did very well in both evaluations. The Academy evaluation suggested that Rolf Nevanlinna -institute might assume a new national role by helping to coordinate an international visitor program in mathematics. The Institute made an initiative to launch such a program, and as of writing this, many parties, notably the Academy of Finland, have expressed their interest in providing the required funds.

To sum up, the people at the Institute can be congratulated for a job well done: thanks to your efforts the year 2000 can definitely be considered a success.
 

Lasse Holmström
 
 
 
 

Figure 1: Some vital statistics of the Institute. Upper panel: Annual budget with the share of the University shown separately within the bars and also as a percentage curve. Middle panel: The number of personnel. Lower panel: The annual number of papers.

Figure 2: Funding of Rolf Nevanlinna Institute by sources.

2  Board of Rolf Nevanlinna Institute

3  Staff of Rolf Nevanlinna Institute

Holmström, Lasse, Docent, Director of the Institute
Sarvas, Jukka, Prof., Director of the Institute (on sabbatical leave in 2000)
Hämäläinen, Tarja, Administrative Secretary
Laakso, Pirjo, Administrative Secretary
Taskinen, Matti, Computer Systems Manager
 

Research Staff

Andreev, Andriy, Research Assistant, M.Sc.
Arjas, Elja, Division Head, Professor
Auranen, Kari, Researcher, Ph.D.
Bhattacharjee, Madhuchhanda, Visiting Researcher, Ph.D.
Corander, Jukka, Researcher, Ph.D.
Eerola, Mervi, Researcher, Ph.D.
Ekholm, Anders, Professor (emer.)
Erästö, Panu, Research Assistant, M.Sc.
Gasbarra, Dario, Researcher, Ph.D.
Holmström, Lasse, Division Head, Docent
Hoti, Fabian, Research Assistant, M.Sc.
Härkänen, Tommi, Research Assistant, Phil.Lic.
Jokinen, Jukka, Research Assistant, M.Pol.Sci.
Järvenpää, Seppo, Research Assistant, Phil.Lic.
Kilpikari, Riika, Research Assistant, M.Pol.Sci.
Kurylev, Yaroslav, Visiting Researcher, Ph.D.
Kärkkäinen, Hanni, Research Assistant, M.Sc.
Klemelä, Jussi, Researcher, Ph.D.
Koistinen, Petri, Researcher, Dr. (Tech.)
Korhonen, Pasi, Research Assistant, M.Pol.Sci.
Lassas, Matti, Researcher, Docent
Lukkarinen, Jani, Research Assistant, M.Sc.
O'Hara, Bob, Visiting Researcher, Ph.D.
Ola, Petri, Visiting Researcher, Ph.D.
Onkamo, Päivi, Research Assistant, M.Sc.
Pennanen, Teemu, Researcher, Ph.D.
Ranta, Jukka, Research Assistant, Phil.Lic
Riiali Anne, Research Assistant, M.Sc.
Ripatti, Samuli, Research Assistant, M.Pol.Sci.
Sarvas, Jukka, Division Head, Professor
Sillanpää, Mikko, Researcher, Ph.D.
Similä, Markku, Visiting Research Assistant, M.Sc.
Smolander, Sampo, Research Assistant, M.Sc.
Somersalo, Erkki, Visiting Researcher, Professor
Taskinen, Matti, Research Assistant
Tietäväinen, Pekka, Research Assistant, M.Sc.
Toivonen, Hannu, Visiting Researcher, Ph.D.
Ukkola, Marko, Research Assistant, M.Sc.
Vasko, Kari, Research Assistant, M.Sc.
Vänskä, Simopekka, Research Assistant, Phi.Lic.
Ylä-Oijala, Pasi, Researcher, Ph.D.
Zurk, Lisa, Visiting Professor, Ph.D.
 
 

4  Research Activities

4.1  Biometry

The staff of the Division consisted of its Head, Prof. Elja Arjas, Ph.D., Prof. (emer.) Anders Ekholm, Ph.D. (from August), KTL senior scientists Mervi Eerola, Ph.D., and Kari Auranen, Ph.D. (part time at the Institute), post doctoral fellows Dario Gasbarra, Ph.D., Mikko Sillanpää, Ph.D., and Bob O'Hara, Ph.D. (part time at the Institute), and research assistants Andriy Andreev, M.Sc. (until June) , Tommi Härkänen, Phil.Lic., Riika Kilpikari, M.Pol.Sci., Hanni Kärkkäinen, M.Sc., Päivi Onkamo, M.Sc., Jukka Ranta, Phil.Lic. (part time at the Institute), Anne Riiali, M.Sc. (until October), Samuli Ripatti, M.Pol.Sci., Markku Similä M.Sc. (part time at the Institute), and Sampo Smolander (from August). Following the dissolution of the Division of Algorithmic Data Analysis at the end of 1999, after its Head Hannu Toivonen, Ph.D., had moved to Nokia Research Center, both he and research assistant Kari Vasko, M.Sc., remained linked to the Institute through the Division of Biometry, with Vasko working at the Institute until July.
 

In more detail, the research carried out in the Division can be described as follows:

Methods for mapping quantitative trait loci in plants and animals, and complex disease traits in humans

This research, which is becoming increasingly important among the activities of the Biometry Division, aims at developing novel methods for mapping quantitative trait loci in plant and animal populations, and for mapping multifactorial disease traits in humans. From a methodological perspective, we continued our earlier work based on Bayesian hierarchical models, using Markov chain Monte Carlo methods in the numerical computations. A new method for carrying out linkage analysis based on general pedigrees was presented by Uimari and Sillanpää in [41]. In addition, new work was initiated by Gasbarra, Sillanpää and Arjas, within the MaDaMe Research Programme, with the aim of reconstructing synthetic ancestral pedigrees on the basis of present day nuclear family genotype and phenotype data, and thereby assessing the role of shared (IBD) haplotypes in the study population. Bayesian mapping methods that had previously been developed by Sillanpää and Arjas were now applied, in a joint work with the plant genetics group led by Prof. Outi Savolainen, University of Oulu, to an analysis of the climatic adaptation in Scots pine [9]. In human genetics, several joint papers by researchers of the Division were presented at the Genetic Analysis Workshop 12 on the genetic mapping based on data from stratified or admixed populations [44,,,85,]. In parallel with these Bayesian approaches a novel data mining approach to linkage disequilibrium mapping in human genetics was introduced by Toivonen, Onkamo, and others [21,27,48,83,84,67], in collaboration with Prof. Juha Kere, Finnish Genome Center. This methodology has been further developed by the group, to include quantitative responses and environmental and biological covariates. The work was presented by Onkamo et al. at the 9th Genetic Epidemiology Society meeting (IGES 2000), where it was awarded the prize for the best presentation by an undergraduate student. The hypothesis that transmission distortion of some diabetic alleles would be an important reason for the increasing incidence of Type 1 diabetes, and the connection of this problem to data ascertainment, were discussed in [18].

Algorithmic data analysis

The development of computational methods for modern data analysis, formerly carried out in the Research Division of Algorithmic Data Analysis, was continued in the Division of Biometry. This research covers a spectrum of methods from algorithms for knowledge discovery and data mining to Bayesian tools and models, as well as to their applications on challenging problems in science and engineering. Data mining algorithms were developed by Toivonen et al. for time series segmentation [64] (with Prof. Heikki Mannila, Nokia Research Center) and for learning in first order logic [31] (with Dr. Luc Dehaspe, Katholieke Universiteit Leuven, Belgium). Tools for Bayesian methodology and MCMC computation were developed [47] and applied to organism-based environmental reconstruction by Vasko and Toivonen et al. [22,40,45] (with Dr. Atte Korhola, Department of Ecology and Systematics) and to metapopulation modeling by O'Hara et al. [75] (with Prof. Ilkka Hanski, Department of Ecology and Systematics).

Statistical modeling and analysis of infectious diseases

Auranen used a back-calculation model to relate incidences of clinical Haemophilus influenzae type b (Hib) disease and subclinical Hib infection, with the aid of a covariate model of the duration immunity to disease [2]. Leino et al. modeled the duration of immunity to Hib disease, using a hierarchical growth curve model of Hib serum antibodies, and discussed possible implications of immunization to clinical Hib disease epidemiology [16]. Auranen et al. applied Markov chain Monte Carlo methods in Bayesian data augmentation for longitudinal data on subclinical pneumoccoccal infection in families [3]. Leino et al.  studied the occurrence of subclinical pneumoccoccal infection in young children using a marginal modeling approach to longitudinal data on infection and risk factors [73].

The usefulness of Markov chain Monte Carlo methods in modeling household outbreaks was studied in an international group of infectious disease modelers [17]. As a part of a European project (ESEN), which evaluated the MMR vaccination program in Europe, the future risk of rubella epidemics was predicted with the aid of deterministic infection models [4]. Eerola, Andreev and Gasbarra modeled the interdependence between subclinical infections, antibody dynamics as a response to randomly occurring infections, and the consequent risk of ear infection by Bayesian data augmentation [61]. Eerola is a member of the FINOM study group at KTL which has studied the efficacy of new pneumococcal conjugate vaccines in a large clinical trial [32]. Eerola and Elena Moltchanova, together with a group of immunologists at KTL, developed Bayesian hierarchical mixture models for evaluating protective immunity in animal models.

A hidden Markov model for predicting meningococcal epidemics in a structured population was studied by Ranta [80]. The model assumes that summary data on the number of disease cases and/or the number of asymptomatic carriers of the bacteria in each subpopulation is partially observed in consecutive time intervals. A related problem of early detection of poliovirus from sewage water specimens was studied by using simulations, under different scenarios, and then comparing this approach to a standard surveillance method under the same scenario [79].

Multivariate and event history models in biometry

Ekholm et al. [5] presented models for longitudinal, or otherwise clustered, binary data which combine regression models for the univariate responses with association models of the subunit responses. A number of mechanisms generating association between subunit responses were formulated and explicit formulas for the likelihood function of the parameters of the combined regression and association models were derived. The theory developed in this paper was applied in [63] to a longitudinal data set, collected by KTL, on bacterial carriage of infants, screened nine times between ages 2 and 18 months. Andreev in his thesis [56], and Arjas and Andreev [1], considered different aspects, including model assessment and the possibility of drawing causal conclusions, of predictive Bayesian inference arising in the context of recurrent events, using acute ear infections in small children as an example. Härkänen et al. [19] considered a multivariate event history model for describing dental caries and failures of permanent teeth, using a nonparametric Bayesian approach in the statistical inference and applying Markov chain Monte Carlo integration in the numerical computations. This model was later applied in a comparative study, considering differences, and trends, in dental caries in three age cohorts and in both sexes [66].

Ripatti, whose Ph.D. work is supervised by Prof. Juni Palmgren from the University of Stockholm, studied the estimation of multivariate frailty models in [20]. This work was motivated by several applied problems, including modeling of hip prosthesis survival, risk of Alzheimer's disease in Swedish twins, and a greenhouse experiment studying lifetimes of cut roses. In addition, Ripatti considered simulation based maximum likelihood estimation in structural equation models [68] and geographical mapping of alcohol related mortality in Finland [74].

Gasbarra and Karia [6] presented a new method for nonparametric Bayesian intensity estimation in a competing risks situation, using failure data arising form the use of intrauterine devices as an example. Arjas wrote two review articles on the use of event history models in a social sciences context, one being concerned with models in discrete time [29] and the other with the possibility of drawing causal conclusions from a statistical analysis of aggregate level data [28].

Spatial and environmental modeling and inference

Ranta continued his earlier work (with Prof. Antti Penttinen, University of Jyväskylä) on spatial models of disease risk with GIS- and register -based data on grid cell specific counts. Specifically, the applications considered were childhood diabetes (Type I) in Finland [19,38], and cancer incidence around a known source of exposure [11]. Riiali, also in collaboration with Penttinen, considered suitability mapping of lichens and spatial modeling of association between lichen species in a forest area [76,77,78].

As a joint effort between the Finnish Institute of Marine Research and the Division of Biometry, and financed through the MaDaMe Research Programme, a new project was started with the aim of developing statistical tools for retrieving parameters of interest from C-band SAR images of sea ice. As a first step of this project, one- dimensional HUTSCAT-scatterometer data with good ground truth validation was examined. A Bayesian hierarchical model was constructed in Similä et al. [86], with the goal of making statistical inferences about the underlying ice surface roughness and about the types of sea ice present.

The geometric shape of coniferous tree crowns can be modeled by using hierarchical structural models, from the level of needles and shoots to that of the canopies. Smolander and Stenberg, continuing their work reported in [39], develop in a MaDaMe Research Programme funded project models for describing the crown structure and its optical properties in conifers, for the purpose of then considering applications to remote sensing, production ecology and global carbon budget research.
 
 

4.2  Mathematical Methods of Information Technology

Last year the staff of the Division consisted of its Head Dr. Lasse Holmström, post-doctoral researchers Jussi Klemelä, Ph.D., and Petri Koistinen, Dr. (Tech.), and research assistants Panu Erästö, M.Sc., and Fabian Hoti, M.Sc. Jussi Klemelä started a research visit at the University of Heidelberg in August.

Computational complexity in kernel estimation

Publication [7] deals with computationally efficient kernel density estimation. A rigorous mathematical analysis of an estimation method based on data prebinning was carried out. Precise formulas for the asymptotic error and the computational complexity of the estimator were derived. Fabian Hoti worked on his licentiate's thesis that considers kernel regression with binned data.

Nonparametric function estimation

In [10] the problem of estimating a probability density function from directional data was considered. Rates of convergence of the kernel estimator were derived and it was proved that asymptotically the plug-in method is as good as using the asymptotically optimal deterministic smoothing parameter sequence.

In [70] lower bounds for the asymptotic minimax risk were given when estimating a density function from directional data. These lower bounds complement the upper bounds obtained in [10]. The lower bounds were formulated in a general setting so that the Euclidean results follow as a special case.

Estimation of the location of a mode of a multivariate density function was studied in [69]. A method for choosing the smoothing parameter of the kernel estimator was presented which has the property that the maximizer of the corresponding kernel estimator converges at the optimal rate to the location of the mode of the density function simultaneously over a scale of smoothness classes of density.

Statistical software and visualization

In [24] a software package written for the XploRe statistical software system was described that implements Classification and Regression Trees (CART). A method of reducing density estimation to regression estimation was also discussed. In [71] a method for the visualization of multivariate density estimates was presented. The method is based on forming a tree from the disconnected parts of the level sets of the estimate. An application to cluster analysis was presented.

Pattern recognition

Complexity in pattern recognition may arise for example from high dimensionality, massive data set size, complicated non-Gaussian structure in data, or from a large number of classes. Various ways to alleviate the ensuing problems were discussed in [43]. Panu Erästö worked on his licentiate's thesis that examines statistical learning theory and the support vector method for pattern classification.

Application to Holocene climate reconstruction

In [65] and [12] we considered the problem of inferring summer temperature trends in Lapland during the past 10 000 years. Temperature reconstruction is based on modeling the relationship between past temperature variation and diatom fossil records obtained from lake sediments in northern Fennoscandia. A new calibration method based on local linear regression was proposed in [65]. The article [12] used multi-scale smoothing to infer trends present at different time scales. This work was also recognized in the national media when the largest Finnish newspaper Helsingin Sanomat described the research involved in a full-page article on June 17, 2000.
 
 

4.3  Mathematical Theory and Applications of Electromagnetic Fields

Inverse boundary and scattering problems

The project on geometrical methods in anisotropic inverse problems was started as part of the work of the research consortium for Anisotropic inverse problems with biomedical applications funded by the Academy of Finland MaDaMe Research Programme. Research on geometrical inverse problems was published in articles [15,14,25,33,35] by Matti Lassas. Kenrick Bingham studied these problems for his Ph.D. thesis. Research on other biomedical inverse problems was also started in the research project on X-ray tomography carried out in collaboration with Instrumentarium Imaging Inc., Invers Inc., University of Kuopio, and Helsinki University of Technology. The geometric inverse conductivity problem was studied by Simopekka Vänskä in his licentiate's thesis [58]. Inverse problems on scattering were studied in [30]. For inverse obstacle scattering problems the linear sampling method was studied by Simopekka Vänskä for this Ph.D. thesis and by Pekka Tietäväinen for his licentiate's thesis.

Absorbing boundary conditions

In computational electromagnetic and acoustic scattering, the exterior domains can be modeled as a bounded domain with an absorbing boundary condition like the Perfectly Matched Layer (PML) condition. In the PML method the computational domain is encircled with a non-reflecting artificial sponge layer. The mathematical theory of these boundary layers was studied by Matti Lassas and Erkki Somersalo in [34]. The numerical application of the PML method was studied by Seppo Järvenpää for his Ph.D. thesis.

Electromagnetic field computing

Microstrip field computing was studied by Matti Taskinen, Pasi Ylä-Oijala, Seppo Järvenpää and Jukka Sarvas [26], and contract research with Aplac Solutions Co continued. The one- and two-layered media Green's functions were extended to the multilayered case with general three dimensional sources [88]. The methods developed were applied to the computation of scattering parameters of three dimensional microstrip structures [50]. Research on fast EM field computing methods was also continued. Jukka Sarvas studied the multilevel fast multipole method in [82], and Marko Ukkola studied the fast volume integral method with FFT for his licentiate's thesis.

Electron multipacting analysis in particle accelerators

The Institute has, in collaboration with DESY (Hamburg, Germany), been studying numerical methods for finding multipacting resonances and methods for analyzing different ways to suppress multipacting in particle accelerators. Pasi Ylä-Oijala, Marko Ukkola, Seppo Järvenpää, Jani Lukkarinen and Jukka Sarvas have participated in this research. The software package for multipacting simulations in axially symmetric structures was extended by adding a 2D FEM electromagnetic field solver [87] and a MATLAB graphical user interface [89]. The package was applied to multipacting in SNS (Spallation Neutron Source) cavities [54,60], and to input couplers [55] designed at the Thomas Jefferson National Accelerator Laboratory. Furthermore, numerical simulations of multipacting suppression in ceramic windows using DC voltage were carried out [42,51].

Magnetostatic and electric field computing

Thin plate magnetostatic field computing with RWG basis functions was studied by Seppo Järvenpää and Jukka Sarvas. The results obtained were applied in collaboration with Neuromag Co to the design of a magnetically shielded room for MEG instrumentation [52]. The collaboration project with the Finnish naval Research Institute on computing electric and magnetic fields of electrodes in a shallow water domain was completed by Simopekka Vänskä [53].

Other activities

The Academy of Finland MALU research project on developing and compiling exercise projects in applied mathematics was continued and completed by Simopekka Vänskä and Jani Lukkarinen [59].
 
 

5  Doctoral Training, Lectures, Seminars, and Workshops

The following lecture series were taught by the Institute: xx¯ 1.  Bayes Models and Data Analysis (E. Arjas)
2.  Electromagnetic and Acoustic Wave Propagation and Scattering (L. Zurk)
3.  Estimation of Functions (L. Holmström)
4.  Fast Multipole Methods in Electromagnetic Field Computing (J. Sarvas)
5.  Mathematical and Statistical Methods for Genetic Analysis (E. Arjas)
6.  Statistical Pattern Recognition (P. Koistinen).

Research seminars: xx¯ 1.  Bayesian Inference (E. Arjas)
2.  Electromagnetism (M. Lassas, P. Ola, and J. Sarvas)
3.  Learning Systems (L. Holmström)
4.  Statistics in Genetics (E. Arjas)
5.  Stochastic Modeling and Data Analysis (E. Arjas, H. Haario, L. Holmström,
 and E. Saksman)

During the fall term the Institute organized, under the auspices of StatNet, a lecture series on ``Bayesian Statistics: Theory and Practice'' given by Professor Anthony O'Hagan (University of Sheffield).

In the summer the Institute provided three summer training positions in its research projects and offered a summer seminar to support the work of the trainees.
 
 

6  Visitors, Visits and Conferences

6.1  Visitors at the Institute

Leonhard Knorr-Held, 11-14 Feb., Imperial College, United Kingdom.
 

Nina Schulman, 6-10 March, Agrifood Research Finland.
 

Jose M. Bernado, 27 March, University of Valencia, Spain.
 

Frank Dudbridge, 8-9 May, University of Cambridge, United Kingdom.
 

Sarah Nutland, 8-9 May, University of Cambridge, United Kingdom.
 

Geert Molenberghs, 20-22 June, University of Limburg, Belgium.
 

Lisa Zurk, 1 Aug.-31 Dec.,MIT Lincoln Laboratory, USA.
 

Anthony O'Hagan, 20-25 Aug., University of Sheffield, United Kingdom.
 

Anthony O'Hagan, 10-16 Sep., University of Sheffield, United Kingdom.
 

Madhuchanda Bhattacharjee, 6 Nov.-1 Dec., Indian Reserve Bank, India.
 

6.2  Visits Abroad

6.2.1  Research Visits

Matti Lassas, Mathematisches Forschungsinstitut Oberwolfach, Germany, 2-21 Jan.
 

Samuli Ripatti, University of Stockholm, Sweden, 17 Jan.-4 Feb.
 

Päivi Onkamo, University of Cambridge, United Kingdom, 1-6 Feb.
 

Matti Lassas, University of Loughborough, UK, 7-27 Feb. and 10-22 July.
 

Samuli Ripatti, Karolinska Institutet, Sweden, 16-17 March.
 

Samuli Ripatti, Karolinska Institutet, Sweden, 27 March-2 April.
 

Päivi Onkamo, University of Cambridge, United Kingdom, 31 March-6 April.
 

Sampo Smolander, Boston University, USA, 1-15 April.
 

Samuli Ripatti, Karolinska Institutet, Sweden, 22-26 May.
 

Jussi Klemelä, Ruprecht-Karls-Universität Heidelberg, Germany, 1 Aug.-31 Dec.
 

Sampo Smolander, Swedish University of Agricultural Sciences (SLU), Sweden, 21-31 Aug.
 

Dario Gasbarra, University of Kiev, Ukraine, 2-15 Sep.
 

Anders Ekholm, Southampton University, UK, 19-24 Sep.
 

Samuli Ripatti, Karolinska Institutet, Sweden, 22-27 Sep.
 

Jukka Jokinen, Southampton University, UK, 1 Oct.-31 Dec.
 

Sampo Smolander, Boston University, USA, 2-12 Nov.
 

Pasi Ylä-Oijala, Jefferson National Accelerator Laboratory, USA, 4-8 Dec.

6.2.2  Conferences, Workshops, Meetings

Anne Riiali, Third French-Danish workshop on spatial statistics and image analysis in biology, Luminy, France, 7-10 March, talk, ``Bayesian biogeographical mapping of lichen species on trees''.
 

Samuli Ripatti, Karolinska Institutet, Sweden, 16 March, talk, ``Modelling age at onset in twin/family studies. Frailty models''.
 

Jukka Sarvas, University of Illinois, ECE seminar, Urbana-Champaign, IL, USA, 18 March, invited talk, ``Electromagnetic field computing''.
 

Kari Auranen, The Second International Symposium on Pneumococci and Pneumococcal Diseases, Sun City, South Africa, March, talk, ``Modelling transmission of pneumococcal carriage in families''.
 

Kari Auranen, The Second International Symposium on Pneumococci and Pneumococcal Diseases, Sun City, South Africa, March, poster, ``Role of family acquisition in nasopharyngeal Pnc carriage in families''.
 

Kari Auranen, The Second International Symposium on Pneumococci and Pneumococcal Diseases, Sun City, South Africa, March, poster, ``Effect of maternal vaccination on pneumococcal carriage in Filipino infants''.
 

Elja Arjas, German Open Conference on Probability and Statistics, Hamburg, Germany, 21-24 March, invited talk, ``Reliability Assessment and Predictive Inference: Some Thoughts on the Foundations''.
 

Dario Gasbarra, National graduate course in Bioinformatics, Göteborg, Sweden, 23-26 March.
 

Lasse Holmström, INTERFACE 2000, The 32nd Symposium in the Interface: Computing Science and Statistics. Modeling Earth's systems: Physical to Infrastructural., New Orleans, Louisiana, USA, 5-8 April, invited talk, ``Using smoothing to reconstruct the holocene temperature in Lapland''.
 

Panu Erästö, INTERFACE 2000, The 32nd Symposium in the Interface: Computing Science and Statistics. Modeling Earth's systems: Physical to Infrastructural., New Orleans, Louisiana, USA, 5-8 April.
 

Petri Koistinen, INTERFACE 2000, The 32nd Symposium in the Interface: Computing Science and Statistics. Modeling Earth's systems: Physical to Infrastructural., New Orleans, Louisiana, USA, 5-8 April.
 

Dario Gasbarra and Elja Arjas, Workshop in Bioinformatics and Statistical Genetics, Göteborg, Sweden, 9-13 May.
 

Mikko Sillanpää, Workshop in Bioinformatics and Statistical Genetics, Göteborg, Sweden, 9-13 May, invited talk, ``Bayesian QTL mapping in inbred and outbred experimental designs''.
 

Lasse Holmström, 5th World Congress of the Bernoulli Society for Mathematical Statistics and Probability and 63rd Meeting of the Institute of Mathematical Statistics, Guanajuato, Mexico, 15-20 May, invited talk, ``Classification of Complex Data''.
 

Fabian Hoti, 5th World Congress of the Bernoulli Society for Mathematical Statistics and Probability and 63rd Meeting of the Institute of Mathematical Statistics, Guanajuato, Mexico, 15-20 May.
 

Petri Koistinen, 5th World Congress of the Bernoulli Society for Mathematical Statistics and Probability and 63rd Meeting of the Institute of Mathematical Statistics, Guanajuato, Mexico, 15-20 May.
 

Kari Auranen, EPIET (European Programme for Intervention Epidemiology Training), Vaccine Module, Glasgow, Great Britain, 20-27 May, invited talk, ``Theory of infectious disease modelling''.
 

Kari Auranen, Advanced Vaccinology Course, Merieux Institute, Annecy, France, 29 May-3 June, invited talk, ``Use of modelling as a tool to predict future events''.
 

Elja Arjas, HSSS Workshop: Bias reduction and confidence estimation in complex models, Lillesand, Norway, 1-3 June, invited talk, ``How many are there?''.
 

Riika Kilpikari, 18th Nordic Conference on Mathematical Statistics, Grimstad, Norway, 4-8 June.
 

Elja Arjas, 18th Nordic Conference on Mathematical Statistics, Grimstad, Norway, 5-8 June, invited discussion on ``Statistical issues in fish stock assessments'' by Stratis Gavaris, and participation in the closing panel discussion of the session ``Large structured models in applied sciences: challenge to statistics''.
 

Samuli Ripatti, International Biometric Conference, Berkeley, California, USA, 4 July, talk, ``Three state frailty model for Alzheimer's disease in Swedish twins''.
 

Jukka Sarvas, PIERS 2000, Boston, MA, USA, 5-14 July, talk, ``Computing the scattering matrix for a microstrip structure''.
 

Kari Auranen, Conference of the International Biometric Society, Berkeley, California, USA, July, talk, ``Backcalculating the age-specific incidence rate of subclinical Haemophilus influenzae type b (Hib) infection''.
 

Anne Riiali, International Conference on Spatial Statistics in the Agro-, Bio- and Geosciences, Freiberg, 19-22 July, talk, ``Bayesian mapping of lichen species on trees''.
 

Simopekka Vänskä, Inverse Problems - Summer School and Conference, Edinburgh, Great Britain, 23 July-6 Aug.
 

Kari Auranen, 40th Interscience Conference on Antimicrobial Agents and Chemotherapy, Toronto, Canada, Sep., poster, ``Antibody response to an elevenvalent pneumococcal conjugate vaccine in Filipino, Finnish and Israeli infants''.
 

Jukka Jokinen and Anders Ekholm, International Conference of the Royal Statistical Society, Reading, UK, 13-15 Sep., poster, ``Combining marginal logistic regression with a model for temporal association''.
 

Samuli Ripatti, Föreningen för Medicinsk Statistik, höstmöte 2000, Lund, Sweden, 22 Sep., invited talk, ``Modelling onset of Alzheimer's disease and death in Swedish twins''.
 

Mikko Sillanpää, Riika Kilpikari, and Samuli Ripatti, 12th Genetic Analysis Workshop, San Antonio, Texas, USA, 23-26 Oct., poster, ``Bayesian association mapping for quantitative traits in recently admixed populations''.
 

Riika Kilpikari, Päivi Onkamo, and Pekka Uimari, 12th Genetic Analysis Workshop, San Antonio, Texas, USA, 23-26 Oct., poster, ``Quantitative trait linkage analysis of GAW 12 simulated data with isolated and general populations''.
 

Kari Auranen, Highly Structured Stochastic Systems (HSSS), closing workshop, Luminy, France, 13-18 Nov., invited discussion on ``Parameters of interest for the control of epidemics in heterogeneous populations'' by Niels Becker.
 

Elja Arjas, Models and inference in HSSS: recent developments and perspectives, Luminy, France, 13-18 Nov., invited discussion on ``Causal inference using influence diagrams: the problem of partial compliance'' by Phil Dawid.
 

Riika Kilpikari, Workshop in Statistical Genetics, Stockholm, Sweden, 16-18 Nov., invited talk, ``Bayesian association mapping for quantitative traits in recently admixed populations''.
 

Päivi Onkamo, Workshop in Statistical Genetics, Stockholm, Sweden, 16-18 Nov., invited talk, ``Haplotype pattern mining extended to accommodate non-genetic covariates and quantitative phenotypes''.
 

Samuli Ripatti, Workshop in Statistical Genetics, Stockholm, Sweden, 16-18 Nov., talk, ``Joint modeling of genetic association and population stratification using latent class models''.
 

Jukka Jokinen, Second Seattle Symposium in Biostatistics: Analysis of Correlated Data, Seattle, USA, 20-21 Nov.
 

Dario Gasbarra, University of Rome, ``La Sapienza'', Rome, Italy, 23 Dec., talk, ``On particle filtering''.
 

Elja Arjas, Recent Developments in Statistics and Probability and Their Applications, New Delhi, India, 30 Dec.2000-2 Jan. 2001, invited talk, ``Modelling dependence by using latent variables: a simple example from repairable systems''.
 
 

7  National Activities of the Institute

xx¯ 1.  Graduate program in Industrial Mathematics.
2.  Functional and Differential Equations in the Complex Domain, Ilomantsi, 31 July-3 Aug.

Publishing Eukleides   The Eukleides news bulletin of the Finnish Mathematical Society was edited and published by the Institute. The editor-in-chief was Jussi Pehkonen. Fifteen issues of Eukleides were edited in 2000.

Doctoral thesis prize   In 2000 the annual Rolf Nevanlinna Institute Prize for an outstanding doctoral thesis in mathematics was awarded to Dr. Jari Kortelainen from the Lappeenranta University of Technology for his thesis ``A Topological Approach to Fuzzy Sets''.
 
 

Papers

Refereed journals

[1]

E. Arjas and A. Andreev. Predictive inference, causal reasoning, and model assessment in nonparametric Bayesian analysis: a case study. Lifetime Data Analysis, 6:187-205, 2000.

[2]

K. Auranen. Back-calculating the age-specific indicence of recurrent subclinical haemophilus influenzae type b infection. Statistics in Medicine, 19:281-296, 2000.

[3]

K. Auranen, E. Arjas, T. Leino, and A. K. Takala. Transmission of pneumococcal carriage in families: a latent Markov process model for binary longitudinal data. Journal of the American Statistical Association, 95:1044-1053, 2000.

[4]

W. J. Edmunds, O. G. Van Heijden, M. Eerola, and N. J. Gay. Modelling rubella in Europe. Epidemiology and Infection, 125:617-634, 2000.

[5]

A. Ekholm, J. W. McDonald, and P. W. F. Smith. Association models for a multivariate binary response. Biometrics, 56:712-718, 2000.

[6]

D. Gasbarra and S. R. Karia. Analysis of competing risks by using Bayesian smoothing. Scandinavian Journal of Statistics, 27:605-617, 2000.

[7]

L. Holmström. The error and the computational complexity of a multivariate binned kernel density estimator. Journal of Multivariate Analysis, 72(2):264-309, 2000.

[8]

T. Härkänen, J. I. Virtanen, and E. Arjas. Caries on permanent teeth: A nonparametric Bayesian analysis. Scandinavian Journal of Statistics, 27:577-588, 2000.

[9]

P. Hurme, M. J. Sillanpää, E. Arjas, T. Repo, and O. Savolainen. Genetic basis of climatic adaptation in Scots pine by Bayesian quantitative trait locus analysis. Genetics, 156:1309-1322, 2000.

[10]

J. Klemelä. Estimation of densities and derivatives of densities with directional data. Journal of Multivariate Analysis, 73(1):18-40, 2000.

[11]

E. Kokki, J. Ranta, A. Penttinen, E. Pukkala, and J. Pekkanen. Small area estimation of cancer incidence around a known source of exposure using fine resolution data. Journal of Occupational and Environmental Medicine, 58:315-320.

[12]

A. Korhola, J. Weckström, L. Holmström, and P. Erästö. A quantitative Holocene climatic record from diatoms in northern Fennoscandia. Quaternary Research, 54:284-294, 2000.

[13]

P. A. Korhonen, T. Loeys, E. Goetghebeur, and J. Palmgren. Vitamin a and infant mortality: beyond intention-to-treat in a randomized trial. Lifetime Data Analysis, 6:107-121, 2000.

[14]

Y. Kurylev and M. Lassas. The Abel-Lidskii basis property in a nonselfadjoint inverse problem. J. Math. Sci., 102(4):4237-4257, 2000. (Translation of Zap. Nauchn. Sem. (POMI) 250 (1998), 161-190, In Russian).

[15]

Y. Kurylev and M. Lassas. Gelf'and inverse problem for a quadratic operator pensil. Journal of Functional Analysis, 176:247-263, 2000.

[16]

T. Leino, K. Auranen, P. H. Mäkelä, H. Käyhty, and A. K. Takala. Dynamics of natural immunity caused by subclinical infections, case study on haemophilus influenzae type b (Hib). Epidemiology and Infection, 125:583-591, 2000.

[17]

P. D. O'Neill, D. J. Balding, N. G. Becker, M. Eerola, and D. Mollison. Analyses of infectious disease data from household outbreaks by markov chain monte carlo methods. Journal of the Royal Statistical Society. Ser.C. Applied statistics, 49:517-542, 2000.

[18]

J. Pitkäniemi, P. Onkamo, E. Arjas, E. Tuomilehto-Wolf, and J. Tuomilehto. Estimation of transmission probabilities in families ascertained through a proband with variable age-at-onset disease: application to the HLA A,B and DR loci in Finnish families with type 1 diabetes. Human Heredity, 50:308-317, 2000.

[19]

J. Ranta and A. Penttinen. Probabilistic small area risk assessment using gis-based data: a case study on Finnish childhood diabetes. Statistics in Medicine, 19:2345-2359, 2000.

[20]

S. Ripatti and J. Palmgren. Estimation of multivariate frailty models using penalized likelihood. Biometrics, 56:1016-1022, 2000.

[21]

H. T. T. Toivonen, P. Onkamo, K. Vasko, V. Ollikainen, P. Sevon, H. Mannila, M. Herr, and J. Kere. Data mining applied to linkage disequilibrium mapping. American Journal of Human Genetics, 67:133-145, 2000.

[22]

K. Vasko, H. T. T. Toivonen, and A. Korhola. A Bayesian multinomial gaussian response model for organism-based environmental reconstruction. Journal of Paleolimnology, 24:243-250, 2000.

[23]

P. Ylä-Oijala. Comparison of boundary integral formulations for electromagnetic field computation in axisymmetric resonators. Journal of Electromagnetic Waves and Applications, 14:767-793, 2000.

 

 
 
 

Refereed edited works and conference proceedings

[24]

J. Klemelä, S. Klinke, and H. Sofyan. Classification and regression trees. In W. Härdle, Z Hlávka, and S. Klinke, editors, XploRe - Application Guide, pages 281-304. Springer, 2000.

[25]

Y. Kyrylev and M. Lassas. Hyperbolic inverse problem with data on a part of the boundary. Differential equations and mathematical physics. In AMS/IP Stud. Adv. Math., 16, pages 259-272, Birmingham, AL, 2000. American Mathematical Society.

[26]

J. Sarvas, M. Taskinen, and S. Järvenpää. Computing the scattering matrix for a microstrip structure with coaxial ports. In AMS/IP Stud. Adv. Math., 16. Proceedings of PIERS 2000 (Progress in Electromagnetics Research Symposium), pages 5-14, Cambridge, Massachusetts, USA, 2000.

[27]

H. T. T. Toivonen, P. Onkamo, K. Vasko, V. Ollikainen, P. Sevon, H. Mannila, and J. Kere. Gene mapping by haplotype pattern mining. IEEE International Symposium on Bio-Informatics and Biomedical Engineering (BIBE 2000), proceedings. - Los Alamitos (CA) : IEEE Computer Society, pages pp. 99-108, Arlington, Virginia, USA, November 2000.

 

 
 
 

Accepted for Publication

[28]

E. Arjas. Causal analysis and statistics: a social sciences perspective. European Sociological Review. Accepted for publication.

[29]

E. Arjas. Longitudinal data: Event history analysis in discrete time. Entry, International Encyclopedia of the Social and Behavioral Sciences. Accepted for publication.

[30]

M. Cheney, D. Isaacson, and M. Lassas:. Optimal acoustic measurements. SIAM Journal on Appiled Mathematics. Accepted for publication.

[31]

Luc Dehaspe and H. T. T. Toivonen. Discovery of relational association rules. In Relational Data Mining: Inductive Logic Programming for Knowledge Discovery in Databases. Springer-Verlag. Accepted for publication.

[32]

J. Eskola, T. Kilpi, A. Palmu, J. Jokinen, M. Eerola, J. Haapakoski, E. Herva, A. Takala, H. Käyhty, P. Karma, R. Kohberger, G. Siber, and P.H. Mäkelä. Efficacy of a sevenvalent pneumococcal polysaccharide - CRM197 conjugate vaccine against serotype-specific, culture-confirmed pneumococcal acute otitis media in infants and children. New England Journal of Medicine. Accepted for publication.

[33]

A. Katchalov and M. Lassas. Gaussian beams and inverse boundary spectral problems. In Springer Lecturene notes. Springer. Accepted for publication.

[34]

M. Lassas and E. Somersalo. Analysis of the PML equations in general convex geometry. Proceedings of Royal Society of Edinburgh Series A. Accepted for publication.

[35]

M. Lassas and G. Uhlmann. Determining Riemannian manifold from boundary measurements. Annales Scientifiques de l' Ecole Normale Superieure. Accepted for publication.

[36]

P. Martikainen, E. Lahelma, and J. Virtamo S. Ripatti. The contribution of smoking on educational differentials in lung cancer mortality. International Journal of Epidemiology. Accepted for publication.

[37]

A. Riiali, A. Penttinen, and M. Kuusinen. Bayesian mapping of lichen growing on trees. Biometrical Journal. Accepted for publication.

[38]

M. Rytkönen, J. Ranta, J. Tuomilehto, and M. Karvonen. Bayesian analysis of geographical variation in the incidence of type i diabetes in Finland. Diabetologia. Accepted for publication.

[39]

S. Smolander and P. Stenberg. A method for estimating light interception by a conifer shoot. Tree Physiology. Accepted for publication.

[40]

H. T. T. Toivonen, H. Mannila, A. Korhola, and H. Olander. Applying Bayesian statistics to organism-based environmental reconstruction. Ecological Applications. Accepted for publication.

[41]

P. Uimari and M. J. Sillanpää. A Bayesian oligogenic analysis of quantitative and qualitative traits in general pedigrees. Genetic Epidemiology. Accepted for publication.

[42]

P. Ylä-Oijala and M. Ukkola. Suppressing electron multipacting in ceramic windows by DC bias. Nuclear Instruments and Methods in Physics Research, Section A. Accepted for publication.

 

 
 
 

Other articles

[43]

L. Holmström, P. Koistinen, F. Hoti, and P. Erästö. Classification of Complex Data. In Year 2000, 5th World Congress of the Bernoulli Society for Mathematical Statistics and Probability and 63rd Meeting of the Institute of Mathematical Statistics. Progrman, Abstracts and Directory of Participants, page 76, Guanajuato, Mexico, 2000. Invited paper.

[44]

R. Kilpikari, P. Onkamo, and P. Uimari. Quantitative trait linkage analysis of gaw12 simulated data with isolated and general populations. In Genetic Analysis Workshop 12 (GAW12), pages 228-232, San Antonio, TX, October 2000.

[45]

A. Korhola, H. Mannila, H. T. T. Toivonen, and K. Vasko. Reconstructing past climate from organism-based fossil assemblages by applying Bayesian statistics. In CSC News, pages 20-22, October 2000.

[46]

S. Ripatti, J. Pitkäniemi, and M. J. Sillanpää. Joint modeling of genetic association and population stratification. In Genetic Analysis Workshop 12 (GAW12), pages 400-404, San Antonio, TX, October 2000.

[47]

M. Salmenkivi, J. Seppänen, H. T. T. Toivonen, and K. Vasko. Bassist - työkalu bayesilaiseen data-analyysiin. @CSC - Tieteen tietotekniikan uutisia Suomessa 3/2000, pages 19-22, 2000.

[48]

P. Sevon, V. Ollikainen, P. Onkamo, H. T. T. Toivonen, H. Mannila, and J. Kere. Mining the associations between genetic marker data and a phenotype including covariates. In Genetic Analysis Workshop 12 (GAW12), pages 440-444, San Antonio, TX, October 2000.

[49]

M. J. Sillanpää, R. Kilpikari, and S. Ripatti. Bayesian association mapping for quantitative traits in recently admixed populations. In Genetic Analysis Workshop 12 (GAW12), pages 459-463, San Antonio, TX, October 2000.

[50]

M. Taskinen and P. Ylä-Oijala. 3D microstrip structures with surface integral equation in multilayered case. In Electromagnetics 2000, The 10th National Electromagnetics Meeting, pages 11-12, Helsinki, Finland, 2000. Nokia Research Center.

[51]

P. Ylä-Oijala and M. Ukkola. Suppressing electron multipacting in TTF III cold window by DC bias. Helsinki Institute of Physics, Preprint series HIP-2000-27/TECH edition, 2000.

 

 
 
 

Contract research reports

[52]

S. Järvenpää and J.Sarvas. Magnetostaattinen heijastus ohuesta levystä tehdystä huoneesta. Research Reports C39, Rolf Nevanlinna Institute, Helsinki, 2000.

[53]

S. Vänskä and J.Sarvas. Pohjatopografian vaikutus elektrodiraivaimen muodostamiin kenttiin - ohutlevy approksimaatio. Research Reports E12, Rolf Nevanlinna Institute, Helsinki, 2000.

[54]

P. Ylä-Oijala. Multipacting simulations on mono-cell SNS superconducting cavities. Rolf Nevanlinna Institute, Helsinki, 2000.

[55]

P. Ylä-Oijala and M. Ukkola. Multipacting simulations on the coaxial SNS coupler. Rolf Nevanlinna Institute, Helsinki, 2000.

 

 
 
 

Theses

[56]

A. Andreev. Nonparametric statistical modeling of recurrent events: a Bayesian approach. PhD thesis, University of Helsinki, 2000. Rolf Nevanlinna Institute Reasearch Reports A32. Yliopistopaino, Helsinki. ISBN 952-9528-59-0. ISSN 0787-8338.

[57]

P. A. Korhonen. Accelerated failure time models for non-ignorable non-compliance in randomized studies. PhD thesis, University of Helsinki, 2000. Rolf Nevanlinna Institute Reasearch Reports A31. Yliopistopaino, Helsinki. ISBN 952-9528-57-4. ISSN 0787-8338.

[58]

S. Vänskä. Determining real analytic metrics by boundary measurements. Research Reports C34, Rolf Nevanlinna Institute, 2000. Licentiate's thesis in applied mathematics, University of Helsinki.

 

 
 
 

Technical reports and preprints

[59]

K. Auranen, J. Lukkarinen, J. Seppänen, and S. Vänskä (editors). Sovelletun matematiikan harjoitustöitä - MALU 2002, 2000.

[60]

D. Barni, G. Ciovati, P. Kneisel, C. Pagani, P. Pierini, and P. Ylä-Oijala. A beta = 0.472 cavity for RIA. Technical report, 2000.

[61]

M. Eerola, A. Andreev, and D. Gasbarra. Joint modelling of recurrent infections and immune response by Bayesian data augmentation. Submitted for publication, 2000.

[62]

A. Ekholm. Johdatus uskottavuuspäättelyyn. 2 painos. Helsinkin Yliopisto, Tilastotieteen laitos, Opetusmoniste 1/2000, 2000.

[63]

A. Ekholm, J. Jokinen, and T. Kilpi. Combining marginal logistic regression with a model for temporal association. Submitted for publication, 2000.

[64]

J. Himberg, K. Korpiaho, H. Mannila, J. Tikanmaki, and H. T. T. Toivonen. Time-series segmentation for context recognition in mobile devices. Submitted for publication, 2000.

[65]

L. Holmström, P. Erästö, P. Koistinen, J. Weckström, and A. Korhola. Using smoothing to reconstruct the Holocene temperature in Lapland. In Proceedings of INTERFACE 2000. The 32nd Symposium in the Interface: Computing Science and Statistics. Modeling the Earth's Systems: Physical to Infrastructural, New Orleans, Lousiana, 2000. Invited paper. To appear.

[66]

T. Härkänen, M. A. Larmas, J. I. Virtanen, and E. Arjas. Applying modern survival analysis methods to longitudinal dental caries studies. Submitted for publication, 2000.

[67]

P. Kauppi, K. Lindblad-Toh, P. Sevon, H. T. T. Toivonen, J. D. Rioux, A. Villapakkam, L. A. Laitinen, J. Kere, T.J. Hudson, and T. Laitinen. A second generation association study on the 5q31 cytokine gene cluster and IL4RA gene in asthma. Submitted for publication, 2000.

[68]

K.Larsen and S. Ripatti. Simulation-based maximum likelihood estimation in linear latent variable models with normal and ordinal outcomes. Technical Report 00/6, Department of Biostatistics, University of Copenhagen, Copenhagen, 2000.

[69]

J. Klemelä. Adaptive estimation of the location of the mode of a multivariate density. Submitted for publication, 2000.

[70]

J. Klemelä. Lower bounds for the asymptotic minimax risk with spherical data. Submitted for publication, 2000.

[71]

J. Klemelä. Visualisation of multivariate density estimates with density trees. Research Reports A33, Rolf Nevanlinna Institute, 2000.

[72]

J. Klemelä, S. Klinke, and H. Sofyan. Classification and regression trees. Discussion Paper 62, Sonderforschungbereich 373, Humboldt Universität, Berlin, 2000.

[73]

T. Leino, K. Auranen, J. Jokinen, M. Leinonen, P. Tervonen, and A. K. Takala. Pneumococcal carriage in children during their first two years; important role of family exposure. Submitted for publication, 2000.

[74]

P. Mäkelä, S. Ripatti, and T. Valkonen. Alue-erot miesten alkoholikuolleisuudessa. Submitted for publication, 2000.

[75]

B. O'Hara, E. Arjas, H. Toivonen, and I. Hanski. Bayesian analysis of metapopulation data. Manuscript, 2000.

[76]

A. Penttinen, F. Divino, and A. Riiali. Hierarchical Bayesian model for spatial small-area association of two species. Manuscript, 2000.

[77]

A. Penttinen, F. Divino, and A. Riiali. Spatial hierarchical Bayesian models in ecological applications. Submitted for publication, 2000.

[78]

A. Penttinen, A. Riiali, and M. Kuusinen. Bayesian modelling of biogeographical association in an inhomogeneous environment. Submitted for publication, 2000.

[79]

J. Ranta, T. Hovi, and E. Arjas. Poliovirus surveillance by examining sewage water specimens. studies on detection probability using simulation models. Submitted for publication, 2000.

[80]

J. Ranta, P. H. Mäkelä, and E. Arjas. Predicting meningococcal disease outbreaks in structured populations. Submitted for publication, 2000.

[81]

S. Ripatti, J. Pitkäniemi, and M. J. Sillanpää. Joint modeling of genetic association and population stratification using latent class models. Submitted for publication, 2000.

[82]

J. Sarvas. Performing interpolation and anterpolation entirely by fast fourier transform in the 3-D multilevel fast multipole algorithm. Research Reports No. CCEM-13-00, Center for Computational Electromagnetics, University of Illinois, Urbana, USA, June 2000.

[83]

P. Sevon, V. Ollikainen, P. Onkamo, H. T. T. Toivonen, H. Mannila, and J. Kere. Mining associations between genetic markers, phenotypes and covariates. Submitted for publication, 2000.

[84]

P. Sevon, H. T. T. Toivonen, and V. Ollikainen. TreeDT: Gene mapping by tree disequilibrium test. Submitted for publication, 2000.

[85]

M. J. Sillanpää, R. Kilpikari, S. Ripatti, P. Onkamo, and P. Uimari. Bayesian association mapping for quantitative traits in mixture of two populations. Submitted for publication, 2000.

[86]

M. Similä, E. Arjas, M. Mäkynen, and M. Hallikainen. Bayesian classification model for sea ice roughness from scatterometer data. Submitted for publication, 2000.

[87]

P. Ylä-Oijala and S. Järvenpää. Finite element method for the electromagnetic field computation in cylindrically symmetric RF structures. Submitted for publication, 2000.

[88]

P. Ylä-Oijala, M. Taskinen, and J. Sarvas. Multilayered media green's functions for MPIE with general electric and magnetic sources by the hertz potential approach. Submitted for publication, 2000.

[89]

P. Ylä-Oijala, M. Ukkola, S. Järvenpää, and J. Lukkarinen. Multipac 2.0 - multipacting simulation toolbox with 2D FEM field solver and MATLAB graphical user interface. Research Reports C35, Rolf Nevanlinna Institute, 2000.