Context-specific independence in graphical models

The theme of this thesis is context-speci c independence in graphical models. Considering a system of stochastic variables it is often the case that the variables are dependent of each other. This can, for instance, be seen by measuring the covariance between a pair of variables. Using graphical models, it is possible to visualize the dependence structure found in a set of stochastic variables. Using ordinary graphical models, such as Markov networks, Bayesian networks, and Gaussian graphical models, the type of dependencies that can be modeled is limited to marginal and conditional (in)dependencies. The models introduced in this thesis enable the graphical representation of context-speci c independencies, i.e. conditional independencies that hold only in a subset of the outcome space of the conditioning variables. In the articles included in this thesis, we introduce several types of graphical models that can represent context-speci c independencies. Models for both discrete variables and continuous variables are considered. A wide range of properties are examined for the introduced models, including identi ability, robustness, scoring, and optimization. In one article, a predictive classi er which utilizes context-speci c independence models is introduced. This classi er clearly demonstrates the potential bene ts of the introduced models. The purpose of the material included in the thesis prior to the articles is to provide the basic theory needed to understand the articles.

Some advances in mathematical models for preference relations

Preference relations, and their modeling, have played a crucial role in both social sciences and applied mathematics. A special category of preference relations is represented by cardinal preference relations, which are nothing other than relations which can also take into account the degree of relation. Preference relations play a pivotal role in most of multi criteria decision making methods and in the operational research. This thesis aims at showing some recent advances in their methodology. Actually, there are a number of open issues in this field and the contributions presented in this thesis can be grouped accordingly. The first issue regards the estimation of a weight vector given a preference relation. A new and efficient algorithm for estimating the priority vector of a reciprocal relation, i.e. a special type of preference relation, is going to be presented. The same section contains the proof that twenty methods already proposed in literature lead to unsatisfactory results as they employ a conflicting constraint in their optimization model. The second area of interest concerns consistency evaluation and it is possibly the kernel of the thesis. This thesis contains the proofs that some indices are equivalent and that therefore, some seemingly different formulae, end up leading to the very same result. Moreover, some numerical simulations are presented. The section ends with some consideration of a new method for fairly evaluating consistency. The third matter regards incomplete relations and how to estimate missing comparisons. This section reports a numerical study of the methods already proposed in literature and analyzes their behavior in different situations. The fourth, and last, topic, proposes a way to deal with group decision making by means of connecting preference relations with social network analysis.

Characteristic Approach to Predicting Analysts’ Forecast Errors in the Finnish Market

Extensive literature shows that analysts’ forecasts and recommendations are often biased. Thus, it is important for the financial market to be able to recognize this bias to be able to correctly valuate public companies. This thesis uses characteristic approach, which was introduced by So (2013, pp. 615-640), to forecast analysts’ forecast errors and tests if predictable forecast error is fully incorporated into share prices. Data is collected of listed Finnish companies. Thesis’ timeframe spans over ten years from 2004 to 2013 consisting of 788 firm-years. Although there is earlier evidence that the characteristic approach is able to predict analysts’ forecast errors, no support for this is found in the Finnish market. This thesis contributes to the current knowledge by showing that the characteristic approach does not work universally as such but requires development to work especially in the smaller markets.

Monte Carlo Methods in Parameter Estimation of Nonlinear Models

Yksi keskeisimmistä tehtävistä matemaattisten mallien tilastollisessa analyysissä on mallien tuntemattomien parametrien estimointi. Tässä diplomityössä ollaan kiinnostuneita tuntemattomien parametrien jakaumista ja niiden muodostamiseen sopivista numeerisista menetelmistä, etenkin tapauksissa, joissa malli on epälineaarinen parametrien suhteen. Erilaisten numeeristen menetelmien osalta pääpaino on Markovin ketju Monte Carlo -menetelmissä (MCMC). Nämä laskentaintensiiviset menetelmät ovat viime aikoina kasvattaneet suosiotaan lähinnä kasvaneen laskentatehon vuoksi. Sekä Markovin ketjujen että Monte Carlo -simuloinnin teoriaa on esitelty työssä siinä määrin, että menetelmien toimivuus saadaan perusteltua. Viime aikoina kehitetyistä menetelmistä tarkastellaan etenkin adaptiivisia MCMC menetelmiä. Työn lähestymistapa on käytännönläheinen ja erilaisia MCMC -menetelmien toteutukseen liittyviä asioita korostetaan. Työn empiirisessä osuudessa tarkastellaan viiden esimerkkimallin tuntemattomien parametrien jakaumaa käyttäen hyväksi teoriaosassa esitettyjä menetelmiä. Mallit kuvaavat kemiallisia reaktioita ja kuvataan tavallisina differentiaaliyhtälöryhminä. Mallit on kerätty kemisteiltä Lappeenrannan teknillisestä yliopistosta ja Åbo Akademista, Turusta.

Origins of production errors and significance of employee empowerment in reducing production error amount in sheet metal fabricating industry

The market place of the twenty-first century will demand that manufacturing assumes a crucial role in a new competitive field. Two potential resources in the area of manufacturing are advanced manufacturing technology (AMT) and empowered employees. Surveys in Finland have shown the need to invest in the new AMT in the Finnish sheet metal industry in the 1990's. In this run the focus has been on hard technology and less attention is paid to the utilization of human resources. In manymanufacturing companies an appreciable portion of the profit within reach is wasted due to poor quality of planning and workmanship. The production flow production error distribution of the sheet metal part based constructions is inspectedin this thesis. The objective of the thesis is to analyze the origins of production errors in the production flow of sheet metal based constructions. Also the employee empowerment is investigated in theory and the meaning of the employee empowerment in reducing the overall production error amount is discussed in this thesis. This study is most relevant to the sheet metal part fabricating industrywhich produces sheet metal part based constructions for electronics and telecommunication industry. This study concentrates on the manufacturing function of a company and is based on a field study carried out in five Finnish case factories. In each studied case factory the most delicate work phases for production errors were detected. It can be assumed that most of the production errors are caused in manually operated work phases and in mass production work phases. However, no common theme in collected production error data for production error distribution in the production flow can be found. Most important finding was still that most of the production errors in each case factory studied belong to the 'human activity based errors-category'. This result indicates that most of the problemsin the production flow are related to employees or work organization. Development activities must therefore be focused to the development of employee skills orto the development of work organization. Employee empowerment gives the right tools and methods to achieve this.

Puumassapesurin suotautumisolojen kartoitus ja simulointi

Työn tavoitteena oli tutkia sellumassan suotautumiseen liittyviä tekijöitä Ahlstrom Machineryn Drum Displacer™ -puumassapesurissa (DD-pesuri). Teoriaosassa tarkasteltiin aluksi suotautumisen teoriaa kuitu-vesi-suspensiossa, minkä jälkeen esiteltiin suotautumisnopeuteen vaikuttavia fysikaalisia ja kemiallisia vaikutusmekanismeja. Seuraavaksi kuvattiin massan pesun yleisiä perusteita sekä teoriaa puumassapesureissa. Lopuksi tarkasteltiin pesurien kytkeytymistä muuhun kuitulinjaan sekä prosessista johtuvia pesun toiminnan ulkoisia häiriötekijöitä. Kokeellisen osan aluksi tarkasteltiin paine- ja lämpötilamittauksien avulla massapesurissa vallitsevia prosessioloja. Mittaustulosten perusteella pumppausolot pesurin suodoslinjoissa ovat vaikeahkot ja häiriötilanteita voi esiintyä, mutta käytäntö on osoittanut tästä olevan vain harvoin haittaa prosessin toiminnalle. Pesureissa toteutuneet syrjäytysnopeudet laskettiin ja niitä verrattiin syrjäytystestien antamiin tuloksiin. Kuitulinjasta riippuen testin vastaavuus tehdasprosessiin vaihteli suuresti. Syrjäytystesteillä kokeiltiin myös tehdasprosesseissa usein esiintyvien muuttujien vaikutusta sellukakun syrjäytettävyyteen. Kakun paksuus ja syrjäytyslämpötila vaikuttivat syrjäytysnopeuteen Darcyn lain mukaisesti. Alipaineen massakakun alapuolella havaittiin huonontavan syrjäytysnopeutta verrattuna tilanteeseen, jossa kakun alla vallitsi ilmanpaine. Tämä havainto on selvästi ristiriidassa suotautumisen teorian kanssa. Massakakun muodostumis-pH osoittautui ratkaisevaksi lopulliselle syrjäytysnopeudelle, sillä alkalisissa oloissa muodostetun kuitukakun syrjäytysnopeus ei enää parantunut happamalla syrjäytysnesteellä. Happamissa oloissa muodostetun kakun syrjäytysnopeus oli alkalisista parempi, mutta se alkoi hitaasti alentua, kun syrjäytysneste vaihtui alkaliseen. Massan laimentaminen ennen syrjäytystä alkalisella tehdassuodoksella puhtaan veden sijasta alensi ligniinipitoisella massalla lopullista syrjäytysnopeutta. Shirato-Tillerin mallilla ja Jönssonin staattisella mallilla simuloitiin numeerisesti syrjäytystestiä kahdessa eri pH:ssa, ja simulointituloksia verrattiin vastaavissa oloissa tehtyihin syrjäytystesteihin. Shirato-Tillerin mallin antamien syrjäytysnopeuksien havaittiin olevan lähellä syrjäytystestien nopeuksia, kun Jönssonin mallin antamat tulokset jäivät huomattavasti testituloksia alemmiksi. Herkkyystarkastelussa havaittiin mallien olevan varsin herkkiä parametrien virheille. Hajonta vaadittavien kuituparametrien määrityksissä ja menetelmien työläys rajoittavat numeerisen simuloinnin käytettävyyttä, sillä kuituparametrien määrityksen vaatima työmäärä on ainakin toistaiseksi syrjäytystestiä suurempi. Lopuksi todettiin, että oikeiden syrjäytysolosuhteiden käyttö on ensiarvoisen tärkeää oikeiden tulosten saamiseksi sekä kokeellisessa että numeerisessa simuloinnissa. Nykyinen syrjäytystestilaitteisto on pienin muutoksin käyttökelpoinen, kun massan testaus prosessioloissa tulee rutiininomaiseksi.

Suomalaisten Internet-kauppojen palvelutaso

Työn tavoitteena oli selvittää Suomalaisten Internet-kauppojen palvelutasoa. Ensin tehtiin tutkimus, jossa mitattiin eri muuttujia, jotka vaikuttavat kaupan palvelutasoon. Internet-kaupasta ostettiin tuote ja kun tuote oli saapunut, se palautettiin takaisin kauppiaalle. Tutkimusosuus päättyi siihen, kun kauppias oli palauttanut rahat asiakkaan tilille. Tutkimuksen jälkeen kerätty data analysoitiin ja etsittiin mahdollisia ongelmia koko ostoprosessista. Ongelmakohtiin tehtiin vaihtoehtoisia ratkaisumalleja, joiden avulla pyritään löytämään paras toimintatapa kussakin tilanteessa. Ongelmina esille tulivat koodausvirheet kauppojen sivuilla, kuolleiden linkkien määrä Solo-torin portaalissa, kauppiaan osoitetietojen puuttuminen kaupan sivuilta ja kuluttajansuojalain kiertäminen tuotteen palautusoikeuden osalta. Pääratkaisuvaihtoehdot ongelmien osalta ovat parempi kuluttajansuojalakien noudattaminen tai toisaalta viranomaisten taholta tarkempi valvonta ja toisaalta Solo-torin linkkien tarkistus tasaisin väliajoin ja osoitetietojen päivitys Internet-kauppojen sivuille.

Mechanistic population Models in Biology: Model Derivation and Application

In general, models of ecological systems can be broadly categorized as ’top-down’ or ’bottom-up’ models, based on the hierarchical level that the model processes are formulated on. The structure of a top-down, also known as phenomenological, population model can be interpreted in terms of population characteristics, but it typically lacks an interpretation on a more basic level. In contrast, bottom-up, also known as mechanistic, population models are derived from assumptions and processes on a more basic level, which allows interpretation of the model parameters in terms of individual behavior. Both approaches, phenomenological and mechanistic modelling, can have their advantages and disadvantages in different situations. However, mechanistically derived models might be better at capturing the properties of the system at hand, and thus give more accurate predictions. In particular, when models are used for evolutionary studies, mechanistic models are more appropriate, since natural selection takes place on the individual level, and in mechanistic models the direct connection between model parameters and individual properties has already been established. The purpose of this thesis is twofold. Firstly, a systematical way to derive mechanistic discrete-time population models is presented. The derivation is based on combining explicitly modelled, continuous processes on the individual level within a reproductive period with a discrete-time maturation process between reproductive periods. Secondly, as an example of how evolutionary studies can be carried out in mechanistic models, the evolution of the timing of reproduction is investigated. Thus, these two lines of research, derivation of mechanistic population models and evolutionary studies, are complementary to each other.

Statistical analysis of survey-based event history data with application to modeling of unemployment duration

Longitudinal surveys are increasingly used to collect event history data on person-specific processes such as transitions between labour market states. Surveybased event history data pose a number of challenges for statistical analysis. These challenges include survey errors due to sampling, non-response, attrition and measurement. This study deals with non-response, attrition and measurement errors in event history data and the bias caused by them in event history analysis. The study also discusses some choices faced by a researcher using longitudinal survey data for event history analysis and demonstrates their effects. These choices include, whether a design-based or a model-based approach is taken, which subset of data to use and, if a design-based approach is taken, which weights to use. The study takes advantage of the possibility to use combined longitudinal survey register data. The Finnish subset of European Community Household Panel (FI ECHP) survey for waves 1–5 were linked at person-level with longitudinal register data. Unemployment spells were used as study variables of interest. Lastly, a simulation study was conducted in order to assess the statistical properties of the Inverse Probability of Censoring Weighting (IPCW) method in a survey data context. The study shows how combined longitudinal survey register data can be used to analyse and compare the non-response and attrition processes, test the missingness mechanism type and estimate the size of bias due to non-response and attrition. In our empirical analysis, initial non-response turned out to be a more important source of bias than attrition. Reported unemployment spells were subject to seam effects, omissions, and, to a lesser extent, overreporting. The use of proxy interviews tended to cause spell omissions. An often-ignored phenomenon classification error in reported spell outcomes, was also found in the data. Neither the Missing At Random (MAR) assumption about non-response and attrition mechanisms, nor the classical assumptions about measurement errors, turned out to be valid. Both measurement errors in spell durations and spell outcomes were found to cause bias in estimates from event history models. Low measurement accuracy affected the estimates of baseline hazard most. The design-based estimates based on data from respondents to all waves of interest and weighted by the last wave weights displayed the largest bias. Using all the available data, including the spells by attriters until the time of attrition, helped to reduce attrition bias. Lastly, the simulation study showed that the IPCW correction to design weights reduces bias due to dependent censoring in design-based Kaplan-Meier and Cox proportional hazard model estimators. The study discusses implications of the results for survey organisations collecting event history data, researchers using surveys for event history analysis, and researchers who develop methods to correct for non-sampling biases in event history data.