Indirect likelihood inference (revised)

Standard indirect Inference (II) estimators take a given finite-dimensional statistic, Z_{n} , and then estimate the parameters by matching the sample statistic with the model-implied population moment. We here propose a novel estimation method that utilizes all available information contained in the distribution of Z_{n} , not just its first moment. This is done by computing the likelihood of Z_{n}, and then estimating the parameters by either maximizing the likelihood or computing the posterior mean for a given prior of the parameters. These are referred to as the maximum indirect likelihood (MIL) and Bayesian Indirect Likelihood (BIL) estimators, respectively. We show that the IL estimators are first-order equivalent to the corresponding moment-based II estimator that employs the optimal weighting matrix. However, due to higher-order features of Z_{n} , the IL estimators are higher order efficient relative to the standard II estimator. The likelihood of Z_{n} will in general be unknown and so simulated versions of IL estimators are developed. Monte Carlo results for a structural auction model and a DSGE model show that the proposed estimators indeed have attractive finite sample properties.

Estudi per a la construcció de l'EDAR mitjançant filtres verds de la Nou de Gaià (Tarragonès)

El tractament de les aigües en nuclis menors de 2000 habitants es troba pendent de completar per part de l’Agència Catalana de l’Aigua més concretament al corresponent Pla de Sanejament d’Aigües Residuals Urbanes (PSARU). El nucli de La Nou de Gaià (al Tarragonès) es troba pendent de la construcció de la corresponent instal·lació de sanejament, projectada al 2007. Alternativament a les depuradores tradicionals basades en l’ús de formigó (o materials alternatius) i en la despesa elèctrica per assegurar una aeració i una evacuació dels fangs generats, existeixen tecnologies “toves”. Aquestes tecnologies, també conegudes com a “verdes”, es basen en imitar els sistemes naturals maximitzant el seu potencial d’autodepuració. A grans trets existeixen dos formes de depurar les aigües de forma ecològica”: llacunatges (existeix una capa d’aigua lliure) i filtres verds. El present estudi es basa en l’aplicació de filtres verds de morfologia vertical i flux subsuperficial, plantat amb canyes dels generes Scirpus o Phragmites. El resultat han estat 4 bases de 35*35 per a tractar un cabal de 150 m3/d i una població equivalent de 1272.

La adolescencia: retos para la investigación y para la sociedad europea de cara al sigol XXI

A lo largo de la última década, la adolescencia ha sido un tema dediscusión política en distintos espacios europeos al más alto nivel. En unasociedad aceleradamente cambiante se percibe que la adecuada socializaciónde las generaciones más jóvenes constituye un reto socio-históricoque nos afecta a todos. Los cambios en que estamos sumergidos son tanplurales (demográficos, sociales, tecnológicos, económicos, políticos,etc.) que generan un amplísimo frente de nuevos dilemas éticos. La opiniónde los ciudadanos de la Unión Europea se muestra preocupada pornuevos valores y destaca la preferencia por la responsabilidad en coherenciacon dicha situación cambiante. Todo este macrocontexto psicosocialviene planteando nuevos retos teóricos y de investigación a la comunidadcientífica. De hecho las ciencias humanas y sociales han empezadoa desarrollar nuevas líneas de investigación para comprender mejor lasnuevas relaciones entre adultos y adolescentes y las nuevas culturas queemergen entre estos últimos, impulsadas por nuevas aspiraciones socialescompartidas por grupos más o menos amplios de la población joven. Eldesarrollo de técnicas e instrumentos que nos permitan comprender mejorla perspectiva del adolescente se hace más evidente si analizamos su relacióncon las nuevas tecnologías de la información y la comunicación. Dichastecnologías comportan nuevos riesgos, pero también nuevas oportunidades,entre las que destaca la posibilidad de establecer nuevas formasde relación. La motivación que muestran los más jóvenes por las nuevastecnologías constituye un gran reto a los investigadores aplicados parasugerir formas de maximizar las potencialidades latentes

Segmentation of Industrial Customers – Case: Finnish Mechanical Engineering Companies

Segmentointi on perinteisesti ollut erityisesti kuluttajamarkkinoinnin työkalu, mutta siirtymä tuotteista palveluihin on lisännyt segmentointitarvetta myös teollisilla markkinoilla. Tämän tutkimuksen tavoite on löytää selkeästi toisistaan erottuvia asiakasryhmiä suomalaisen liikkeenjohdon konsultointiyritys Synocus Groupin tarjoaman case-materiaalin pohjalta. K-means-klusteroinnin avulla löydetään kolme potentiaalista markkinasegmenttiä perustuen siihen, mitkä tarjoamaelementit 105 valikoitua suomalaisen kone- ja metallituoteteollisuuden asiakasta ovat maininneet tärkeimmiksi. Ensimmäinen klusteri on hintatietoiset asiakkaat, jotka laskevat yksikkökohtaisia hintoja. Toinen klusteri koostuu huolto-orientoituneista asiakkaista, jotka laskevat tuntikustannuksia ja maksimoivat konekannan käyttötunteja. Tälle kohderyhmälle kannattaisi ehkä markkinoida teknisiä palveluja ja huoltosopimuksia. Kolmas klusteri on tuottavuussuuntautuneet asiakkaat, jotka ovat kiinnostuneita suorituskyvyn kehittämisestä ja laskevat tonnikohtaisia kustannuksia. He tavoittelevat alempia kokonaiskustannuksia lisääntyneen suorituskyvyn, pidemmän käyttöiän ja alempien huoltokustannusten kautta.

The harm principle and the greatest happiness principle: the missing link

In this article I present a possible solution for the classic problem of the apparent incompatibility between Mill's Greatest Happiness Principle and his Principle of Liberty arguing that in the other-regarding sphere the judgments of experience and knowledge accumulated through history have moral and legal force, whilst in the self-regarding sphere the judgments of the experienced people only have prudential value and the reason for this is the idea according to which each of us is a better judge than anyone else to decide what causes us pain and which kind of pleasure we prefer (the so-called epistemological argument). Considering that the Greatest Happiness Principle is nothing but the aggregate of each person's happiness, given the epistemological claim we conclude that, by leaving people free even to cause harm to themselves, we still would be maximizing happiness, so both principles (the Greatest Happiness Principle and the Principle of Liberty) could be compatible.

MCMC Analysis for Optimization of Stochastic Models

In any decision making under uncertainties, the goal is mostly to minimize the expected cost. The minimization of cost under uncertainties is usually done by optimization. For simple models, the optimization can easily be done using deterministic methods.However, many models practically contain some complex and varying parameters that can not easily be taken into account using usual deterministic methods of optimization. Thus, it is very important to look for other methods that can be used to get insight into such models. MCMC method is one of the practical methods that can be used for optimization of stochastic models under uncertainty. This method is based on simulation that provides a general methodology which can be applied in nonlinear and non-Gaussian state models. MCMC method is very important for practical applications because it is a uni ed estimation procedure which simultaneously estimates both parameters and state variables. MCMC computes the distribution of the state variables and parameters of the given data measurements. MCMC method is faster in terms of computing time when compared to other optimization methods. This thesis discusses the use of Markov chain Monte Carlo (MCMC) methods for optimization of Stochastic models under uncertainties .The thesis begins with a short discussion about Bayesian Inference, MCMC and Stochastic optimization methods. Then an example is given of how MCMC can be applied for maximizing production at a minimum cost in a chemical reaction process. It is observed that this method performs better in optimizing the given cost function with a very high certainty.

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.

Innovation Project Portfolio Management

Companies today are forced to innovate in order to remain within business. Such innovation projects undertaken by the companies are defined in this study as creative ideas which have been managed through “Stage-Gate” innovation process. This process is used to manage innovation projects as they proceed from being newly created to ready for launching/implementing. This has ensured that the companies manage the innovation project right. However, with so many new creative ideas the companies can come up within limited resources, the companies must rely on Innovation Project Portfolio Management (IPPM) to ensure that they are managing only the right innovation projects. Although, there are many tools and techniques available for use within Project Portfolio Management, there is still no consensus on which are the most effective and no standard framework has been established especially for IPPM. Thus, this study proposes a practical framework for which individual innovative organization can follow as a guideline to manage its innovation project portfolio. The study theoretically first addresses the key differences between project portfolio management of innovation projects and other traditional projects - one of which is the stage nature of innovation projects due to their unclear objectives from the beginning compare to clearly established objectives of traditional projects. Secondly, different tools and techniques which can be used are examined based on the three goals of IPPM: (1) Maximizing the Value of Innovation Project Portfolio: Financial Methods, Decision Trees, Scoring Models and Checklists; (2) Balancing Innovation Project Portfolio: Visual Representations; and (3) Aligning Innovation Project Portfolio with Strategy: Bottom-Up (Scoring Models with Strategic Criteria) and Top-Down (Strategic Buckets). Finally, the two approaches in which IPPM can be integrated with Stage-Gate innovation process are discussed: (1) Gates- Dominated; and (2) Portfolio Reviews-Dominated. Practically, this study investigates IPPM of a case organization, and through analysis of the case study results proposes a practical framework for case organization to improve its current management of innovation project portfolio. This framework is then generalized to propose a final practical framework or guideline for which an innovative organization can follow to manage its innovation project portfolio.

Viilunkuivaajan tuotantotehokkuuden kehittäminen

Mekaanisen metsäteollisuuden tuotteiden kiristyvä globaali kilpailu ja jatkuva kustannusten kasvaminen kiristyvillä markkinoilla aiheuttavat jatkuvaa tuotannon tehostamisen tarvetta. Käyntiaikojen maksimoinnilla ja tuotantokoneiden käyttöasteen nostamisella haetaan hintaetua sekä toimituskapasiteetin maksimointia kilpailussa asiakkaista myös tulevaisuutta silmälläpitäen. Vanerin tuotantoprosessiin osaprosessina kuuluva viilun kuivaus on tehtaan tuotannon pullonkaula. Kuivauskoneen tuotannollisten häiriöiden syiden selvittämisillä ja käyntiajan lisäämisellä on suora tuotannollinen vaikutus. Tutkimuksessa löydettiin mahdollisia parannuskohteita tuotantolaitteisiin. Käsiteltävän aiheen ollessa aiemmin kirjallisuudessa käsittelemätön, jouduttiin kehitys- ja parannustoimenpiteissä tekemään suoraviivaisia syy- ja seuraussuhteisiin perustuvia ratkaisuja. Vaihtoehdoista etsittiin toiminnaltaan mahdollisimman yksinkertaisia sekä hankinta- ja käyttökustannuksiltaan kustannustehokkaita ratkaisuja.

Hydramata: Building a Nimble Solution with Hydra to Transcend the Institutional Repository

Presentation at Open Repositories 2014, Helsinki, Finland, June 9-13, 2014

RFID-teknologia kulunvalvonnassa

Tämä kandidaatintutkielma on kirjallisuuskatsaus, joka käsittelee RFID-teknologian hyöndyntämistä kulunvalvonnassa. Työssä perehdytään pintapuolisesti itse teknologiaan, ja luodaan katsaus kulunvalvontaan. Työn pääaihe on kuitenkin kulunvalvonnan ja RFID:n yhdistyminen: miten RFID:tä hyödynnetään kulunvalvonnan toteutuksissa ympäri maailmaa. Työssä tarkastellaan RFID:n vahvuuksia, sekä heikkouksia kulunvalvonnan suhteen. Tämän lisäksi pyritään luomaan kuva nykyisistä ja tulevista implementaatioista. Viimeinen tärkeä työn osa-alue on turvallisuus. RFID:tä käytetään korkeankin turvatason kulunvalvontaratkaisuissa ja tällöin turvallisuuden maksimoiminen on ensiarvoisen tärkeää.

Strategic planning of major disturbance proof network

In the 2000’s Finland suffered from storms that caused long outages in electricity distribution, longest up to two weeks. These major disturbances increased the importance of supply security. In 2013 new Electricity Market Act was announced. It defined maximum duration for outages, 6 h for city plan areas and 36 h for other areas. The aim for this work is to determine required major disturbance proof level for a study area and find tools for prioritizing overhead lines for cabling renovation to improve supply security. Three prioritization methods were chosen to be studied: A: prioritization line sections by customer outage costs they cause, B: maximizing customers major disturbance proof network and C: minimizing excavation costs in medium voltage network. Profitability calculations showed that prioritization method A was the most profitable and C had the weakest profitability. The prioritization method C drove renovation into unreasonable locations in the study area in reliability point of view. Therefore universal rule prioritization methods couldn’t be made from the prioritization methods. This led to the conclusion that every renewing area need to be evaluated in a case by case basis.

Intracranial Nimodipine Implant: Feasibility and Implications for the Treatment of Subarachnoid Hemorrhage – A Pre-Clinical Study

Intracranial aneurysmal subarachnoid hemorrhage (aSAH) is a life-threatening condition requiring immediate neurocritical care. A ruptured aneurysm must be isolated from arterial circulation to prevent rebleeding. Open surgical clipping of the neck of the aneurysm or intra-arterial filling of the aneurysm sack with platinum coils are major treatment strategies in an acute phase. About 40% of the patients suffering from aSAH die within a year of the bleeding despite the intensive treatment. After aSAH, the patient may develop a serious complication called vasospasm. Major risk for the vasospasm takes place at days 5–14 after the primary bleeding. In vasospasm, cerebral arteries contract uncontrollably causing brain ischemia that may lead to death. Nimodipine (NDP) is used to treat of vasospasm and it is administrated intravenously or orally every four hours for 21 days. NDP treatment has been scientifically proven to improve patients’ clinical outcome. The therapeutic effect of L-type calcium channel blocker NDP is due to the ability to dilate cerebral arteries. In addition to vasodilatation, recent research has shown the pleiotropic effect of NDP such as inhibition of neuronal apoptosis and inhibition of microthrombi formation. Indeed, NDP inhibits cortical spreading ischemia. Knowledge of the pathophysiology of the vasospasm has evolved in recent years to a complex entity of early brain injury, secondary injuries and cortical spreading ischemia, instead of being pure intracranial vessel spasm. High NDP levels are beneficial since they protect neurons and inhibit the cortical spreading ischemia. One of the drawbacks of the intravenous or oral administration of NPD is systemic hypotension, which is harmful particularly when the brain is injured. Maximizing the beneficial effects and avoiding systemic hypotension of NDP, we developed a sustained release biodegradable NDP implant that was surgically positioned in the basal cistern of animal models (dog and pig). Higher concentrations were achieved locally and lower concentrations systemically. Using this treatment approach in humans, it may be possible to reduce incidence of harmful hypotension and potentiate beneficial effects of NDP on neurons. Intracellular calcium regulation has a pivotal role in brain plasticity. NDP blocks L-type calcium channels in neurons, substantially decreasing intracellular calcium levels. Thus, we were interested in how NDP affects brain plasticity and tested the hypothesis in a mouse model. We found that NDP activates Brain-derived neurotrophic factor (BDNF) receptor TrkB and its downstream signaling in a reminiscent of antidepressant drugs. In contrast to antidepressant drugs, NDP activates Akt, a major survival-promoting factor. Our group’s previous findings demonstrate that long-term antidepressant treatment reactivates developmental-type of plasticity mechanisms in the adult brain, which allows the remodeling of neuronal networks if combined with appropriate rehabilitation. It seems that NDP has antidepressant-like properties and it is able to induce neuronal plasticity. In general, drug induced neuronal plasticity has a huge potential in neurorehabilitation and more studies are warranted.

Design and Drive-Cycle Based Analysis of Direct-Driven Permanent Magnet Synchronous Machine for a Small Urban Use Electric Vehicle

A direct-driven permanent magnet synchronous machine for a small urban use electric vehicle is presented. The measured performance of the machine at the test bench as well as the performance over the modified New European Drive Cycle will be given. The effect of optimal current components, maximizing the efficiency and taking into account the iron loss, is compared with the simple id=0 – control. The machine currents and losses during the drive cycle are calculated and compared with each other.

Development of an Ecology-Oriented Software-Defined Networking Framework

ICT contributed to about 0.83 GtCO2 emissions where the 37% comes from the telecoms infrastructures. At the same time, the increasing cost of energy has been hindering the industry in providing more affordable services for the users. One of the sources of these problems is said to be the rigidity of the current network infrastructures which limits innovations in the network. SDN (Software Defined Network) has emerged as one of the prominent solutions with its idea of abstraction, visibility, and programmability in the network. Nevertheless, there are still significant efforts needed to actually utilize it to create a more energy and environmentally friendly network. In this paper, we suggested and developed a platform for developing ecology-related SDN applications. The main approach we take in realizing this goal is by maximizing the abstractions provided by OpenFlow and to expose RESTful interfaces to modules which enable energy saving in the network. While OpenFlow is made to be the standard for SDN protocol, there are still some mechanisms not defined in its specification such as settings related to Quality of Service (QoS). To solve this, we created REST interfaces for setting of QoS in the switches which can maximize network utilization. We also created a module for minimizing the required network resources in delivering packets across the network. This is achieved by utilizing redundant links when it is needed, but disabling them when the load in the network decreases. The usage of multi paths in a network is also evaluated for its benefit in terms of transfer rate improvement and energy savings. Hopefully, the developed framework can be beneficial for developers in creating applications for supporting environmentally friendly network infrastructures.