Covariance Matrix Estimation and Mean-Variance Optimization in application to International Diversification

Tutkimus keskittyy kansainväliseen hajauttamiseen suomalaisen sijoittajan näkökulmasta. Tutkimuksen toinen tavoite on selvittää tehostavatko uudet kovarianssimatriisiestimaattorit minimivarianssiportfolion optimointiprosessia. Tavallisen otoskovarianssimatriisin lisäksi optimoinnissa käytetään kahta kutistusestimaattoria ja joustavaa monimuuttuja-GARCH(1,1)-mallia. Tutkimusaineisto koostuu Dow Jonesin toimialaindekseistä ja OMX-H:n portfolioindeksistä. Kansainvälinen hajautusstrategia on toteutettu käyttäen toimialalähestymistapaa ja portfoliota optimoidaan käyttäen kahtatoista komponenttia. Tutkimusaieisto kattaa vuodet 1996-2005 eli 120 kuukausittaista havaintoa. Muodostettujen portfolioiden suorituskykyä mitataan Sharpen indeksillä. Tutkimustulosten mukaan kansainvälisesti hajautettujen investointien ja kotimaisen portfolion riskikorjattujen tuottojen välillä ei ole tilastollisesti merkitsevää eroa. Myöskään uusien kovarianssimatriisiestimaattoreiden käytöstä ei synnytilastollisesti merkitsevää lisäarvoa verrattuna otoskovarianssimatrisiin perustuvaan portfolion optimointiin.

Studies on integrating kinematic design method with mechanical systems simulation techniques

Theultimate goal of any research in the mechanism/kinematic/design area may be called predictive design, ie the optimisation of mechanism proportions in the design stage without requiring extensive life and wear testing. This is an ambitious goal and can be realised through development and refinement of numerical (computational) technology in order to facilitate the design analysis and optimisation of complex mechanisms, mechanical components and systems. As a part of the systematic design methodology this thesis concentrates on kinematic synthesis (kinematic design and analysis) methods in the mechanism synthesis process. The main task of kinematic design is to find all possible solutions in the form of structural parameters to accomplish the desired requirements of motion. Main formulations of kinematic design can be broadly divided to exact synthesis and approximate synthesis formulations. The exact synthesis formulation is based in solving n linear or nonlinear equations in n variables and the solutions for the problem areget by adopting closed form classical or modern algebraic solution methods or using numerical solution methods based on the polynomial continuation or homotopy. The approximate synthesis formulations is based on minimising the approximation error by direct optimisation The main drawbacks of exact synthesis formulationare: (ia) limitations of number of design specifications and (iia) failure in handling design constraints- especially inequality constraints. The main drawbacks of approximate synthesis formulations are: (ib) it is difficult to choose a proper initial linkage and (iib) it is hard to find more than one solution. Recentformulations in solving the approximate synthesis problem adopts polynomial continuation providing several solutions, but it can not handle inequality const-raints. Based on the practical design needs the mixed exact-approximate position synthesis with two exact and an unlimited number of approximate positions has also been developed. The solutions space is presented as a ground pivot map but thepole between the exact positions cannot be selected as a ground pivot. In this thesis the exact synthesis problem of planar mechanism is solved by generating all possible solutions for the optimisation process ¿ including solutions in positive dimensional solution sets - within inequality constraints of structural parameters. Through the literature research it is first shown that the algebraic and numerical solution methods ¿ used in the research area of computational kinematics ¿ are capable of solving non-parametric algebraic systems of n equations inn variables and cannot handle the singularities associated with positive-dimensional solution sets. In this thesis the problem of positive-dimensional solutionsets is solved adopting the main principles from mathematical research area of algebraic geometry in solving parametric ( in the mathematical sense that all parameter values are considered ¿ including the degenerate cases ¿ for which the system is solvable ) algebraic systems of n equations and at least n+1 variables.Adopting the developed solution method in solving the dyadic equations in direct polynomial form in two- to three-precision-points it has been algebraically proved and numerically demonstrated that the map of the ground pivots is ambiguousand that the singularities associated with positive-dimensional solution sets can be solved. The positive-dimensional solution sets associated with the poles might contain physically meaningful solutions in the form of optimal defectfree mechanisms. Traditionally the mechanism optimisation of hydraulically driven boommechanisms is done at early state of the design process. This will result in optimal component design rather than optimal system level design. Modern mechanismoptimisation at system level demands integration of kinematic design methods with mechanical system simulation techniques. In this thesis a new kinematic design method for hydraulically driven boom mechanism is developed and integrated in mechanical system simulation techniques. The developed kinematic design method is based on the combinations of two-precision-point formulation and on optimisation ( with mathematical programming techniques or adopting optimisation methods based on probability and statistics ) of substructures using calculated criteria from the system level response of multidegree-of-freedom mechanisms. Eg. by adopting the mixed exact-approximate position synthesis in direct optimisation (using mathematical programming techniques) with two exact positions and an unlimitednumber of approximate positions the drawbacks of (ia)-(iib) has been cancelled.The design principles of the developed method are based on the design-tree -approach of the mechanical systems and the design method ¿ in principle ¿ is capable of capturing the interrelationship between kinematic and dynamic synthesis simultaneously when the developed kinematic design method is integrated with the mechanical system simulation techniques.

Development of phase retrieval algorithm and techniques for optical spectrum analysis

Coherent anti-Stokes Raman scattering is the powerful method of laser spectroscopy in which significant successes are achieved. However, the non-linear nature of CARS complicates the analysis of the received spectra. The objective of this Thesis is to develop a new phase retrieval algorithm for CARS. It utilizes the maximum entropy method and the new wavelet approach for spectroscopic background correction of a phase function. The method was developed to be easily automated and used on a large number of spectra of different substances.. The algorithm was successfully tested on experimental data.

Computational Techniques for Spectral Image Analysis

Multispectral images are becoming more common in the field of remote sensing, computer vision, and industrial applications. Due to the high accuracy of the multispectral information, it can be used as an important quality factor in the inspection of industrial products. Recently, the development on multispectral imaging systems and the computational analysis on the multispectral images have been the focus of a growing interest. In this thesis, three areas of multispectral image analysis are considered. First, a method for analyzing multispectral textured images was developed. The method is based on a spectral cooccurrence matrix, which contains information of the joint distribution of spectral classes in a spectral domain. Next, a procedure for estimating the illumination spectrum of the color images was developed. Proposed method can be used, for example, in color constancy, color correction, and in the content based search from color image databases. Finally, color filters for the optical pattern recognition were designed, and a prototype of a spectral vision system was constructed. The spectral vision system can be used to acquire a low dimensional component image set for the two dimensional spectral image reconstruction. The data obtained by the spectral vision system is small and therefore convenient for storing and transmitting a spectral image.

Evaluating multidimensional visualization techniques in data mining tasks

Visual data mining (VDM) tools employ information visualization techniques in order to represent large amounts of high-dimensional data graphically and to involve the user in exploring data at different levels of detail. The users are looking for outliers, patterns and models – in the form of clusters, classes, trends, and relationships – in different categories of data, i.e., financial, business information, etc. The focus of this thesis is the evaluation of multidimensional visualization techniques, especially from the business user’s perspective. We address three research problems. The first problem is the evaluation of projection-based visualizations with respect to their effectiveness in preserving the original distances between data points and the clustering structure of the data. In this respect, we propose the use of existing clustering validity measures. We illustrate their usefulness in evaluating five visualization techniques: Principal Components Analysis (PCA), Sammon’s Mapping, Self-Organizing Map (SOM), Radial Coordinate Visualization and Star Coordinates. The second problem is concerned with evaluating different visualization techniques as to their effectiveness in visual data mining of business data. For this purpose, we propose an inquiry evaluation technique and conduct the evaluation of nine visualization techniques. The visualizations under evaluation are Multiple Line Graphs, Permutation Matrix, Survey Plot, Scatter Plot Matrix, Parallel Coordinates, Treemap, PCA, Sammon’s Mapping and the SOM. The third problem is the evaluation of quality of use of VDM tools. We provide a conceptual framework for evaluating the quality of use of VDM tools and apply it to the evaluation of the SOM. In the evaluation, we use an inquiry technique for which we developed a questionnaire based on the proposed framework. The contributions of the thesis consist of three new evaluation techniques and the results obtained by applying these evaluation techniques. The thesis provides a systematic approach to evaluation of various visualization techniques. In this respect, first, we performed and described the evaluations in a systematic way, highlighting the evaluation activities, and their inputs and outputs. Secondly, we integrated the evaluation studies in the broad framework of usability evaluation. The results of the evaluations are intended to help developers and researchers of visualization systems to select appropriate visualization techniques in specific situations. The results of the evaluations also contribute to the understanding of the strengths and limitations of the visualization techniques evaluated and further to the improvement of these techniques.

Kernel-Based Ranking. Methods for Learning and Performance Estimation

Machine learning provides tools for automated construction of predictive models in data intensive areas of engineering and science. The family of regularized kernel methods have in the recent years become one of the mainstream approaches to machine learning, due to a number of advantages the methods share. The approach provides theoretically well-founded solutions to the problems of under- and overfitting, allows learning from structured data, and has been empirically demonstrated to yield high predictive performance on a wide range of application domains. Historically, the problems of classification and regression have gained the majority of attention in the field. In this thesis we focus on another type of learning problem, that of learning to rank. In learning to rank, the aim is from a set of past observations to learn a ranking function that can order new objects according to how well they match some underlying criterion of goodness. As an important special case of the setting, we can recover the bipartite ranking problem, corresponding to maximizing the area under the ROC curve (AUC) in binary classification. Ranking applications appear in a large variety of settings, examples encountered in this thesis include document retrieval in web search, recommender systems, information extraction and automated parsing of natural language. We consider the pairwise approach to learning to rank, where ranking models are learned by minimizing the expected probability of ranking any two randomly drawn test examples incorrectly. The development of computationally efficient kernel methods, based on this approach, has in the past proven to be challenging. Moreover, it is not clear what techniques for estimating the predictive performance of learned models are the most reliable in the ranking setting, and how the techniques can be implemented efficiently. The contributions of this thesis are as follows. First, we develop RankRLS, a computationally efficient kernel method for learning to rank, that is based on minimizing a regularized pairwise least-squares loss. In addition to training methods, we introduce a variety of algorithms for tasks such as model selection, multi-output learning, and cross-validation, based on computational shortcuts from matrix algebra. Second, we improve the fastest known training method for the linear version of the RankSVM algorithm, which is one of the most well established methods for learning to rank. Third, we study the combination of the empirical kernel map and reduced set approximation, which allows the large-scale training of kernel machines using linear solvers, and propose computationally efficient solutions to cross-validation when using the approach. Next, we explore the problem of reliable cross-validation when using AUC as a performance criterion, through an extensive simulation study. We demonstrate that the proposed leave-pair-out cross-validation approach leads to more reliable performance estimation than commonly used alternative approaches. Finally, we present a case study on applying machine learning to information extraction from biomedical literature, which combines several of the approaches considered in the thesis. The thesis is divided into two parts. Part I provides the background for the research work and summarizes the most central results, Part II consists of the five original research articles that are the main contribution of this thesis.

First dice your dill – new methods and techniques in sample handling

This book is dedicated to celebrate the 60th birthday of Professor Rainer Huopalahti. Professor Rainer “Repe” Huopalahti has had, and in fact is still enjoying a distinguished career in the analysis of food and food related flavor compounds. One will find it hard to make any progress in this particular field without a valid and innovative sample handling technique and this is a field in which Professor Huopalahti has made great contributions. The title and the front cover of this book honors Professor Huopahti’s early steps in science. His PhD thesis which was published on 1985 is entitled “Composition and content of aroma compounds in the dill herb, Anethum graveolens L., affected by different factors”. At that time, the thesis introduced new technology being applied to sample handling and analysis of flavoring compounds of dill. Sample handling is an essential task that in just about every analysis. If one is working with minor compounds in a sample or trying to detect trace levels of the analytes, one of the aims of sample handling may be to increase the sensitivity of the analytical method. On the other hand, if one is working with a challenging matrix such as the kind found in biological samples, one of the aims is to increase the selectivity. However, quite often the aim is to increase both the selectivity and the sensitivity. This book provides good and representative examples about the necessity of valid sample handling and the role of the sample handling in the analytical method. The contributors of the book are leading Finnish scientists on the field of organic instrumental analytical chemistry. Some of them are also Repe’ s personal friends and former students from the University of Turku, Department of Biochemistry and Food Chemistry. Importantly, the authors all know Repe in one way or another and are well aware of his achievements on the field of analytical chemistry. The editorial team had a great time during the planning phase and during the “hard work editorial phase” of the book. For example, we came up with many ideas on how to publish the book. After many long discussions, we decided to have a limited edition as an “old school hard cover book” – and to acknowledge more modern ways of disseminating knowledge by publishing an internet version of the book on the webpages of the University of Turku. Downloading the book from the webpage for personal use is free of charge. We believe and hope that the book will be read with great interest by scientists working in the fascinating field of organic instrumental analytical chemistry. We decided to publish our book in English for two main reasons. First, we believe that in the near future, more and more teaching in Finnish Universities will be delivered in English. To facilitate this process and encourage students to develop good language skills, it was decided to be published the book in English. Secondly, we believe that the book will also interest scientists outside Finland – particularly in the other member states of the European Union. The editorial team thanks all the authors for their willingness to contribute to this book – and to adhere to the very strict schedule. We also want to thank the various individuals and enterprises who financially supported the book project. Without that support, it would not have been possible to publish the hardcover book.

Solving large sparse linear systems over the field GF(2)

Nowadays problem of solving sparse linear systems over the field GF(2) remain as a challenge. The popular approach is to improve existing methods such as the block Lanczos method (the Montgomery method) and the Wiedemann-Coppersmith method. Both these methods are considered in the thesis in details: there are their modifications and computational estimation for each process. It demonstrates the most complicated parts of these methods and gives the idea how to improve computations in software point of view. The research provides the implementation of accelerated binary matrix operations computer library which helps to make the progress steps in the Montgomery and in the Wiedemann-Coppersmith methods faster.