Parameter identification in time series models

Time series analysis can be categorized into three different approaches: classical, Box-Jenkins, and State space. Classical approach makes a basement for the analysis and Box-Jenkins approach is an improvement of the classical approach and deals with stationary time series. State space approach allows time variant factors and covers up a broader area of time series analysis. This thesis focuses on parameter identifiablity of different parameter estimation methods such as LSQ, Yule-Walker, MLE which are used in the above time series analysis approaches. Also the Kalman filter method and smoothing techniques are integrated with the state space approach and MLE method to estimate parameters allowing them to change over time. Parameter estimation is carried out by repeating estimation and integrating with MCMC and inspect how well different estimation methods can identify the optimal model parameters. Identification is performed in probabilistic and general senses and compare the results in order to study and represent identifiability more informative way.

Evolutionary Ecology of Mountain Birch in Subarctic Stress Gradients: Interplay of Biotic and Abiotic Factors in Plant-Plant Interactions and Evolutionary Processes

In nature, variation for example in herbivory, wind exposure, moisture and pollution impact often creates variation in physiological stress and plant productivity. This variation is seldom clear-cut, but rather results in clines of decreasing growth and productivity towards the high-stress end. These clines of unidirectionally changing stress are generally known as ‘stress gradients’. Through its effect on plant performance, stress has the capacity to fundamentally alter the ecological relationships between individuals, and through variation in survival and reproduction it also causes evolutionary change, i.e. local adaptations to stress and eventually speciation. In certain conditions local adaptations to environmental stress have been documented in a matter of just a few generations. In plant-plant interactions, intensities of both negative interactions (competition) and positive ones (facilitation) are expected to vary along stress gradients. The stress-gradient hypothesis (SGH) suggests that net facilitation will be strongest in conditions of high biotic and abiotic stress, while a more recent ‘humpback’ model predicts strongest net facilitation at intermediate levels of stress. Plant interactions on stress gradients, however, are affected by a multitude of confounding factors, making studies of facilitation-related theories challenging. Among these factors are plant ontogeny, spatial scale, and local adaptation to stress. The last of these has very rarely been included in facilitation studies, despite the potential co-occurrence of local adaptations and changes in net facilitation in stress gradients. Current theory would predict both competitive effects and facilitative responses to be weakest in populations locally adapted to withstand high abiotic stress. This thesis is based on six experiments, conducted both in greenhouses and in the field in Russia, Norway and Finland, with mountain birch (Betula pubescens subsp. czerepanovii) as the model species. The aims were to study potential local adaptations in multiple stress gradients (both natural and anthropogenic), changes in plant-plant interactions under conditions of varying stress (as predicted by SGH), potential mechanisms behind intraspecific facilitation, and factors confounding plant-plant facilitation, such as spatiotemporal, ontogenetic, and genetic differences. I found rapid evolutionary adaptations (occurring within a time-span of 60 to 70 years) towards heavy-metal resistance around two copper-nickel smelters, a phenomenon that has resulted in a trade-off of decreased performance in pristine conditions. Heavy-metal-adapted individuals had lowered nickel uptake, indicating a possible mechanism behind the detected resistance. Seedlings adapted to heavy-metal toxicity were not co-resistant to others forms of abiotic stress, but showed co-resistance to biotic stress by being consumed to a lesser extent by insect herbivores. Conversely, populations from conditions of high natural stress (wind, drought etc.) showed no local adaptations, despite much longer evolutionary time scales. Due to decreasing emissions, I was unable to test SGH in the pollution gradients. In natural stress gradients, however, plant performance was in accordance with SGH, with the strongest host-seedling facilitation found at the high-stress sites in two different stress gradients. Factors confounding this pattern included (1) plant size / ontogenetic status, with seedling-seedling interactions being competition dominated and host-seedling interactions potentially switching towards competition with seedling growth, and (2) spatial distance, with competition dominating at very short planting distances, and facilitation being strongest at a distance of circa ¼ benefactor height. I found no evidence for changes in facilitation with respect to the evolutionary histories of plant populations. Despite the support for SGH, it may be that the ‘humpback’ model is more relevant when the main stressor is resource-related, while what I studied were the effects of ‘non-resource’ stressors (i.e. heavy-metal pollution and wind). The results have potential practical applications: the utilisation of locally adapted seedlings and plant facilitation may increase the success of future restoration efforts in industrial barrens as well as in other wind-exposed sites. The findings also have implications with regard to the effects of global change in subarctic environments: the documented potential by mountain birch for rapid evolutionary change, together with the general lack of evolutionary ‘dead ends’, due to not (over)specialising to current natural conditions, increase the chances of this crucial forest-forming tree persisting even under the anticipated climate change.

Spatiotemporal Features of Coastal Waters in Southwest Finland

In coastal waters, physico-chemical and biological properties and constituents vary at different time scales. In the study area of this thesis, within the Archipelago Sea in the northern Baltic Sea, seasonal cycles of light and temperature set preconditions for intra-annual variations, but developments at other temporal scales occur as well. Weather-induced runoffs and currents may alter water properties over the short term, and the consequences over time of eutrophication and global changes are to a degree unpredictable. The dynamic characteristics of northern Baltic Sea waters are further diversified at the archipelago coasts. Water properties may differ in adjacent basins, which are separated by island and underwater thresholds limiting water exchange, making the area not only a mosaic of islands but also one of water masses. Long-term monitoring and in situ observations provide an essential data reserve for coastal management and research. Since the seasonal amplitudes of water properties are so high, inter-annual comparisons of water-quality variables have to be based on observations sampled at the same time each year. In this thesis I compare areas by their temporal characteristics, using both inter-annual and seasonal data. After comparing spatial differences in seasonal cycles, I conclude that spatial comparisons and temporal generalizations have to be made with caution. In classifying areas by the state of their waters, the results may be biased even if the sampling is annually simultaneous, since the dynamics of water properties may vary according to the area. The most comprehensive view of the spatiotemporal dynamics of water properties would be achieved by means of comparisons with data consisting of multiple annual samples. For practical reasons, this cannot be achieved with conventional in situ sampling. A holistic understanding of the spatiotemporal features of the water properties of the Archipelago Sea will have to be based on the application of multiple methods, complementing each other’s spatial and temporal coverage. The integration of multi-source observational data and time-series analysis may be methodologically challenging, but it will yield new information as to the spatiotemporal regime of the Archipelago Sea.

Modeling dynamic behavior in tilting pad gas journal bearings

Thisresearch deals with the dynamic modeling of gas lubricated tilting pad journal bearings provided with spring supported pads, including experimental verification of the computation. On the basis of a mathematical model of a film bearing, a computer program has been developed, which can be used for the simulation of a special type of tilting pad gas journal bearing supported by a rotary spring under different loading conditions time dependently (transient running conditions due to geometry variations in time externally imposed). On the basis of literature, different transformations have been used in the model to achieve simpler calculation. The numerical simulation is used to solve a non-stationary case of a gasfilm. The simulation results were compared with literature results in a stationary case (steady running conditions) and they were found to be equal. In addition to this, comparisons were made with a number of stationary and non-stationary bearing tests, which were performed at Lappeenranta University of Technology using bearings designed with the simulation program. A study was also made using numerical simulation and literature to establish the influence of the different bearing parameters on the stability of the bearing. Comparison work was done with literature on tilting pad gas bearings. This bearing type is rarely used. One literature reference has studied the same bearing type as that used in LUT. A new design of tilting pad gas bearing is introduced. It is based on a stainless steel body and electron beam welding of the bearing parts. It has good operation characteristics and is easier to tune and faster to manufacture than traditional constructions. It is also suitable for large serial production.

A Co-Processor Approach for Efficient Java Execution in Embedded Systems

This thesis deals with a hardware accelerated Java virtual machine, named REALJava. The REALJava virtual machine is targeted for resource constrained embedded systems. The goal is to attain increased computational performance with reduced power consumption. While these objectives are often seen as trade-offs, in this context both of them can be attained simultaneously by using dedicated hardware. The target level of the computational performance of the REALJava virtual machine is initially set to be as fast as the currently available full custom ASIC Java processors. As a secondary goal all of the components of the virtual machine are designed so that the resulting system can be scaled to support multiple co-processor cores. The virtual machine is designed using the hardware/software co-design paradigm. The partitioning between the two domains is flexible, allowing customizations to the resulting system, for instance the floating point support can be omitted from the hardware in order to decrease the size of the co-processor core. The communication between the hardware and the software domains is encapsulated into modules. This allows the REALJava virtual machine to be easily integrated into any system, simply by redesigning the communication modules. Besides the virtual machine and the related co-processor architecture, several performance enhancing techniques are presented. These include techniques related to instruction folding, stack handling, method invocation, constant loading and control in time domain. The REALJava virtual machine is prototyped using three different FPGA platforms. The original pipeline structure is modified to suit the FPGA environment. The performance of the resulting Java virtual machine is evaluated against existing Java solutions in the embedded systems field. The results show that the goals are attained, both in terms of computational performance and power consumption. Especially the computational performance is evaluated thoroughly, and the results show that the REALJava is more than twice as fast as the fastest full custom ASIC Java processor. In addition to standard Java virtual machine benchmarks, several new Java applications are designed to both verify the results and broaden the spectrum of the tests.

Evaluating the success of software development projects in Russia, Ukraine, and Belarus

The size and complexity of projects in the software development are growing very fast. At the same time, the proportion of successful projects is still quite low according to the previous research. Although almost every project's team knows main areas of responsibility which would help to finish project on time and on budget, this knowledge is rarely used in practice. So it is important to evaluate the success of existing software development projects and to suggest a method for evaluating success chances which can be used in the software development projects. The main aim of this study is to evaluate the success of projects in the selected geographical region (Russia-Ukraine-Belarus). The second aim is to compare existing models of success prediction and to determine their strengths and weaknesses. Research was done as an empirical study. A survey with structured forms and theme-based interviews were used as the data collection methods. The information gathering was done in two stages. At the first stage, project manager or someone with similar responsibilities answered the questions over Internet. At the second stage, the participant was interviewed; his or her answers were discussed and refined. It made possible to get accurate information about each project and to avoid errors. It was found out that there are many problems in the software development projects. These problems are widely known and were discussed in literature many times. The research showed that most of the projects have problems with schedule, requirements, architecture, quality, and budget. Comparison of two models of success prediction presented that The Standish Group overestimates problems in project. At the same time, McConnell's model can help to identify problems in time and avoid troubles in future. A framework for evaluating success chances in distributed projects was suggested. The framework is similar to The Standish Group model but it was customized for distributed projects.

Characteristics of credit risk assessment methods of major Nordic banks in allocating capital when extending credits to large corporations

Credit risk assessment is an integral part of banking. Credit risk means that the return will not materialise in case the customer fails to fulfil its obligations. Thus a key component of banking is setting acceptance criteria for granting loans. Theoretical part of the study focuses on key components of credit assessment methods of Banks in the literature when extending credits to large corporations. Main component is Basel II Accord, which sets regulatory requirement for credit risk assessment methods of banks. Empirical part comprises, as primary source, analysis of major Nordic banks’ annual reports and risk management reports. As secondary source complimentary interviews were carried out with senior credit risk assessment personnel. The findings indicate that all major Nordic banks are using combination of quantitative and qualitative information in credit risk assessment model when extending credits to large corporations. The relative input of qualitative information depends on the selected approach to the credit rating, i.e. point-in-time or through-the-cycle.

Atomic level phenomena on transition metal surfaces

In this study we discuss the atomic level phenomena on transition metal surfaces. Transition metals are widely used as catalysts in industry. Therefore, reactions occuring on transition metal surfaces have large industrial intrest. This study addresses problems in very small size and time scales, which is an important part in the overall understanding of these phenomena. The publications of this study can be roughly divided into two categories: The adsorption of an O2 molecule to a surface, and surface structures of preadsorbed atoms. These two categories complement each other, because in the realistic case there are always some preadsorbed atoms at the catalytically active surfaces. However, all transition metals have an active d-band, and this study is also a study of the in uence of the active d-band on other atoms. At the rst part of this study we discuss the adsorption and dissociation of an O2 molecule on a clean stepped palladium surface and a smooth palladium surface precovered with sulphur and oxygen atoms. We show how the reactivity of the surface against the oxygen molecule varies due to the geometry of the surface and preadsorbed atoms. We also show how the molecular orbitals of the oxygen molecule evolve when it approaches the di erent sites on the surface. In the second part we discuss the surface structures of transition metal surfaces. We study the structures that are intresting on account of the Rashba e ect and charge density waves. We also study the adsorption of suphur on a gold surface, and surface structures of it. In this study we use ab-initio based density functional theory methods to simulate the results. We also compare the results of our methods to the results obtained with the Low-Energy-Electron-Difraction method.

Orthogonal function method in modelling paper web quality variation

This work is devoted to the analysis of signal variation of the Cross-Direction and Machine-Direction measurements from paper web. The data that we possess comes from the real paper machine. Goal of the work is to reconstruct the basis weight structure of the paper and to predict its behaviour to the future. The resulting synthetic data is needed for simulation of paper web. The main idea that we used for describing the basis weight variation in the Cross-Direction is Empirical Orthogonal Functions (EOF) algorithm, which is closely related to Principal Component Analysis (PCA) method. Signal forecasting in time is based on Time-Series analysis. Two principal mathematical procedures that we used in the work are Autoregressive-Moving Average (ARMA) modelling and Ornstein–Uhlenbeck (OU) process.

The Finnish Paper Workers' Union at a crossroads: Labor union representativeness in a changing environment, 1980-2008

This thesis consists of four articles and an introductory section. The main research questions in all the articles refer to the changes in the representativeness of the Finnish Paper Workers' Union. Representativeness stands for the entire entity of external, internal, legal and reputational factors that enable the labor union to represent its members and achieve its goals. This concept is based on an extensive reading of quantitative and qualitative industrial relations literature, which includes works based on Marxist labor-capital relations (such as Hyman's industrial relations studies), and more recent union density studies as well as gender- and ethnic diversity-based 'union revitalization' studies. Müller-Jentsch's German studies of industrial relations have been of particular importance as well as Streeck's industrial unionism and technology studies. The concept of representativeness is an attempt to combine the insights of these diverse strands of literature and bring the scientific discussion of labor unions back to the core of a union's function: representing its members. As such, it can be seen as a theoretical innovation. The concept helps to acknowledge both the heterogeneity of the membership and the totality of a labor union organization. The concept of representativeness aims to move beyond notions of 'power'. External representativeness can be expressed through the position of the labor union in the industrial relations system and the economy. Internal representativeness focuses on the aspects of labor unions that relate to the function of the union as an association with members, such as internal democracy. Legal representativeness lies in the formal legal position of the union – its rights and instruments. This includes collective bargaining legislation, co-decision rules and industrial conflict legislation. Reputational representativeness is related to how the union is seen by other actors and the general public, and can be approximated using data on strike activity. All these aspects of representativeness are path-dependent, and show the results of previous struggles over issues. The concept of representativeness goes beyond notions of labor union power and symbolizes an attempt to bring back the focus of industrial relations studies to the union's basic function of representing its members. The first article shows in detail the industrial conflict of the Finnish paper industry in 2005. The intended focus was the issue of gender in the negotiations over a new collective agreement, but the focal point of the industrial conflict was the issue of outsourcing and how this should be organized. Also, the issue of continuous shifts as an issue of working time was very important. The drawn-out conflict can be seen as a struggle over principles, and under pressure the labor union had to concede ground on the aforementioned issues. The article concludes that in this specific conflict, the union represented its' female members to a lesser extent, because the other issues took such priority. Furthermore, because of the substantive concessions. the union lost some of its internal representativeness, and the stubbornness of the union may have even harmed the reputation of the union. This article also includes an early version of the representativeness framework, through which this conflict is analyzed. The second article discusses wage developments, union density and collective bargaining within the context of representativeness. It is shown that the union has been able to secure substantial benefits for its members, regardless of declining employment. Collective agreements have often been based on centralized incomes policies, but the paper sector has not always joined these. Attention is furthermore paid to the changing competition of the General Assembly, with a surprisingly strong position of the Left Alliance still. In an attempt to replicate analysis of union density measures, an analysis of sectoral union density shows that similar factors as in aggregate data influence this measure, though – due to methodological issues – the results may not be robust. On this issue, it can be said that the method of analysis for aggregate union density is not suitable for sectoral union density analysis. The increasingly conflict-ridden industrial relations predicted have not actually materialized. The article concludes by asking whether the aim of ever-increasing wages is a sustainable one in the light of the pressures of globalization, though wage costs are a relatively small part of total costs. The third article discusses the history and use of outsourcing in the Finnish paper industry. It is shown using Hyman's framework of constituencies that over time, the perspective of the union changed from 'members of the Paper Workers' Union' to a more specific view of who is a core member of the union. Within the context of the industrial unionism that the union claims to practice, this is an important change. The article shows that the union more and more caters for a core group, while auxiliary personnel is less important to the union's identity and constituencies, which means that the union's internal representativeness has decreased. Maintenance workers are an exception; the union and employers have developed a rotating system that increases the efficient allocation of these employees. The core reason of the exceptional status of maintenance personnel is their high level of non-transferable skills. In the end it is debatable whether the compromise on outsourcing solves the challenges facing the industry. The fourth article shows diverging discourses within the union with regard to union-employer partnership for competitiveness improvements and instruments of local union representatives. In the collective agreement of 2008, the provision regulating wage effects of significant changes in the organization or content of work was thoroughly changed, though this mainly reflected decisions by the Labor Court on the pre-2008 version of the provision. This change laid bare the deep rift between the Social Democratic and Left Alliance (ex-Communist) factions of the union. The article argues that through the changed legal meaning of the provision, the union was able to transform concession bargaining into a basis for partnership. The internal discontent about this issue is nonetheless substantial and a threat to the unity of the union, both locally and at the union level. On the basis of the results of the articles, other factors influencing representativeness, such as technology and EU law and an overview of the main changes in the Finnish paper industry, it is concluded that, especially in recent years, the Finnish Paper Workers' Union has lost some of its representativeness. In particular, the loss of the efficiency of strikes is noted, the compromise on outsourcing which may have alienated a substantial part of the union's membership, and the change in the collective agreement of 2008 have caused this decline. In the latter case, the internal disunion on that issue shows the constraints of the union's internal democracy. Furthermore, the failure of the union to join the TEAM industrial union (by democratic means), the internal conflicts and a narrow focus on its own sector may also hurt the union in the future, as the paper industry in Finland is going through a structural change. None of these changes in representativeness would have been so drastic without the considerable pressure of globalization - in particular changing markets, changing technology and a loss of domestic investments to foreign investments, which in the end have benefited the corporations more than the Finnish employees of these corporations. Taken together, the union risks becoming socially irrelevant in time, though it will remain formally very strong on the basis of its institutional setting and financial situation.

De-escalation of commitment management strategy in IT project management methodology: case of ERP implementation project by IT consulting company

According to several surveys and observations, the percentage of successfully conducted IT projects without over-budgeting and delays in time schedule are extremely low. Many projects also are evaluated as failures in terms of delivered functionality. Nuldén (1996) compares IT projects with bad movies; after watching for 2 hours, one still tries to finish it even though one understands that it is a complete waste of time. The argument for that is 'I've already invested too much time to terminate it now'. The same happens with IT projects: sometimes the company continues wasting money on these projects for a long time, even though there are no expected benefits from these projects. Eventually these projects are terminated anyway, but until this moment, the company spends a lot. The situation described above is a consequence of “escalation of commitment” - project continuation even after a manager receives negative feedback of the project’s success probability. According to Keil and Mähring (2010), even though escalation can occur in any type of project, it is more common among complex technological projects, such as IT projects. Escalation of commitment very often results in runaway projects. In order to avoid it, managers use de-escalation strategies, which allow the resources to be used in more effective. These strategies lead to project termination or turning around, which stops the flow of wasted investments. Numbers of researches explore escalation of commitment phenomena based on experiments and business cases. Moreover, during the last decade several frameworks were proposed for de-escalation strategy. However, there is no evidence of successful implementation of the de-escalation of commitment strategy in the literature. In addition, despite that fact that IT project management methodologies are widely used in the companies, none of them cover the topic of escalation of commitment risks. At the same time, there are no researches proposing the way to implement de-escalation of commitment strategy into the existing project management methodology The research is focused on a single case of large ERP implementation project by the consulting company. Hence, the main deliverables of the study include suggestions of improvement in de-escalation methods and techniques in the project and in the company. Moreover, the way to implement these methods into existing project management methodology and into the company general policies is found.

Entrepreneurial orientation in small and medium-sized enterprises during economic crisis

The economic importance of small and medium-sized enterprises (SMEs) and entrepreneurship has increased significantly in recent decades and entrepreneurial activity and SMEs are deemed vital to economic progress. Therefore, it is justifiable to study how small firms and entrepreneurs can enhance their performance and emergence in the turbulent economic environment. The concept of entrepreneurial orientation (EO) has recently attracted considerable attention in the field of entrepreneurship research. EO generally refers to a firm’s propensity to be innovative, to be proactive and to take risks. A majority of EO studies so far have found that adopting EO associated entrepreneurial behaviors will help firms to create or sustain a high level of performance. This dissertation explores the main drivers and performance implications of EO for SMEs in time of economic crisis. Hence the first objective of this dissertation is to examine the performance implications of EO and to test the role of EO on how firms are treated by the crisis at operative level. The second objective is to expand the prevailing understanding of determinants of EO by exploring the relationship between owner's work related values, attitudes, demographic characteristics, firm’s financial resources and EO. EO was found to be a significant and positive factor behind a firm’s long run growth. Hence it can be said that EO has positive implications for firm performance. But on the other hand, during a time of economic crisis the different dimensions of EO had both positive and negative effects on performance of SMEs. The performance implications varied across different stages of the crisis and were also dependent on what measure was used for measuring the performance. The main drivers of EO in SMEs were the personal work related values of the entrepreneur and his/her prior experience as an entrepreneur. The intrinsic work values related to interest, responsibility, challenge, self-development or intellectual stimulation and values related to status, power, achievement and recognition had a positive effect on the level of EO. On the other hand, extrinsic values related to high income, material possessions, benefits such as generous holidays, job security, and comfort through good working conditions decreased the level of EO