1,721,018 research outputs found
Optimal Design in Hierarchical Random Effect Models for Individual Prediction with Application in Precision Medicine
No full textCorrection to: Optimal Design in Hierarchical Random Effect Models for Individual Prediction with Application in Precision Medicine
No full textDesigns for first-order interactions in paired comparison experiments with two-level factors
No full textFor paired comparison experiments involving options described by a common set of two-level factors a new method for generating exact designs is presented. These designs allow the efficient estimation of main effects and first order interactions and perform better than alternative designs available in the literature
Going Beyond Counting First Authors in Author Co-citation Analysis
Full text linkThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
Variations on the Author
Full text link“Variations on the Author” discusses two of Eduardo Coutinho’s recent films (Um Dia na Vida, from 2010, and Últimas Conversas, posthumously released in 2015) and their contribution to the general question of documentary authorship. The director’s filmography is characterized by a consistent yet self-effacing form of authorial self-inscription: Coutinho often features as an interviewer that rather than express opinions propels discourses; an interviewer that is good at listening. This mode of self-inscription characterizes him as an author who is not expressive but who is nonetheless markedly present on the screen. In Um Dia na Vida, however, Coutinho is completely absent form the image, while Últimas Conversas, on the contrary, includes a confessional prologue that moves the director from the margins to the center of his films. This article examines the ways in which these works stand out in the filmography of a director who offers new insights into the notion of cinematic authorship
Appropriate Similarity Measures for Author Cocitation Analysis
Full text linkWe provide a number of new insights into the methodological discussion about author cocitation analysis. We first argue that the use of the Pearson correlation for measuring the similarity between authors’ cocitation profiles is not very satisfactory. We then discuss what kind of similarity measures may be used as an alternative to the Pearson correlation. We consider three similarity measures in particular. One is the well-known cosine. The other two similarity measures have not been used before in the bibliometric literature. Finally, we show by means of an example that our findings have a high practical relevance.information science;Pearson correlation;cosine;similarity measure;author cocitation analysis
Optimal designs for the prediction in hierarchical random coefficient regression models
Full text linkvon Maryna Pru
Theoretische Grundlagen der partiellen kleinsten Quadrate
Full text linkMagdeburg, Univ., Fak. für Mathematik, Diss., 2014von Hayan Hasa
Optimal designs for paired comparison experiments
Full text linkThe aim of this thesis is to derive optimal designs for linear paired comparison models
with second- or third-order interactions in an analysis of variance setup where the
attributes are qualitative with the same number of levels each.
After the first introductory chapter on the problem and the literature some basic
concepts are presented in the Chapter 2 about paired comparison experiments in the
linear model setup, and particular emphasis is laid on the special case of the part-worth
model in which the influence of the attributes is additive and consists only of main
effects without interactions. The fundamentals and general descriptions of optimal
designs as well as some commonly used optimality criteria are presented in Chapter 3.
In Chapters 4 and 5 results are presented on optimal designs for the part-worth model
and a model with first-order interactions, respectively, which have been known from the
literature and where components of a single attribute are used as building blocks for
the holistic approach. A powerful tool for characterizing the optimal designs in these
models is given by the concept of invariance.
These concepts are extended to linear paired comparison models with second-order
interactions in Chapters 6 and 7 for binary attributes and for attributes with a general
common number of levels, respectively. A general statement on the maximal number of
types of pairs can be formulated for optimal designs, where orbits are specified by the
number of attributes in which the two alternatives differ. While for part-worth (main
effects) models optimal designs consist of those alternatives which differ in all attributes
and for first-order interactions they consist of those pairs of alternatives which differ
in about half of the attributes, respectively, there seems to be no clear general rule in
models with second-order interactions. For models with small profile strengths analytic
results can be obtained for optimal designs while for larger profile strengths optimal
designs have to be determined numerically. Moreover, for binary attributes optimal
designs require two types of pairs in which either all attributes have distinct levels or
approximately half of the attributes are distinct and the other half of the attributes
coincide. For larger number of levels mostly one type of pairs is sufficient. In some
exceptional cases two types of pairs are needed, and only for the full interaction case
all types are required.
In Chapters 8 and 9 these results are extended to paired comparison models with
third-order interactions. For binary attributes two types of pairs have to be considered
for which the numbers of distinct attributes are symmetric with respect to about half of
the profile strength. For larger number of levels again only one type of pair is sufficient
in nearly all cases.
The thesis is concluded with a brief discussion and an outlook on future research.Das Ziel dieser Arbeit ist die Herleitung optimaler experimenteller Designs für Paarvergleichsmodelle
unter Zugrundelegung von Linearen Modellen der Varianzanalyse mit
Wechselwirkungen zweiter bzw. dritter Ordnung. Dabei setzen sich die Alternativen
aus mehreren, die Entscheidungen beeinflussenden Attributen zusammen, die jeweils
auf eine feste Anzahl von Ausprägungen (Stufen) eingestellt werden können.
Nach einem einleitenden Kapitel in die Problemstellung und die Literatur werden
im zweiten Kapitel die grundlegenden Konzepte für Paarvergleiche im Linearen Modell
eingeführt. Dabei wird der klassische Spezialfall des Teilwertmodells, in dem nur
Haupteffekte der Attribute und keine Wechselwirkungen auftreten, gesondert behandelt.
Darauf folgen im drtitten Kapitel grundlegende Erläuterungen zu optimalen Designs
sowie zu üblicherweise verwendeten Optimalitätskriterien.
Die Interaktionsmodelle zweiter und dritter Ordnung, welche das Teilwertmodell
mit Komponenten eines einzelnen Attributs als Bausteine für die Resultate benutzen,
sowie die Interaktionsmodelle erster Ordnung, werden in Kapitel 4 und 5 beschrieben.
Von besonderer Relevanz ist dabei das Konzept der Invarianz.
In den Kapiteln 6 und 7 werden diese Konzepte auf lineare Paarvergleichsmodelle
mit Interaktionen zweiter Ordnung und binären Attributen, sowie Attributen von
identischer Stufenanzahl, erweitert. Hierbei kann ein allgemeingültiges Resultat über die
maximale Anzahl benötigter Typen von Paaren für optimale Designs formuliert werden,
wobei die verschiedenen Typen von Paaren durch die Anzahl von unterschiedlichen
Attributen der Alternativen spezifiziert werden. Dabei bestehen optimale Designs
für Teilwertmodelle aus den Alternativen, in denen sich alle Attribute unterscheiden,
während Interaktionsmodelle erster Ordnung aus Alternativen bestehen, die sich in
ungefähr der Hälfte der Attribute unterscheiden. Im Fall der Interaktionsmodelle
zweiter Ordnung scheint es keine derart allgemeingültige Regel zu geben. Für Modelle
von kleinen Profilstärken können analytische Lösungen für optimale Designs gefunden
werden, während größere Profilstärken numerische Methoden erfordern. Zudem bedürfen
optimale Designs im Fall von binären Attributen zweier Typen von Paaren, in denen
entweder alle Attribute unterschiedliche Stufen haben oder ungefähr jeweils die Hälfte
der Attribute identisch und unterschiedlich sind. Für eine größere Anzahl von Stufen
genügt in der Regel ein Typ von Paaren. In Ausnahmefällen bedarf es zweier Typen von
Paaren, und nur für den Fall vollständiger Interaktion werden alle Typen von Paaren
benötigt.
In den Kapiteln 8 und 9 werden die zuvor beschriebenen Resultate auf Paarvergleichsmodelle
mit Interaktionen der dritten Ordnung erweitert. Für binäre Attribute
müssen wieder zwei Typen von Paaren verwendet werden, für die die Anzahl der
verschiedenen Attribute symmetrisch zur Hälfte der Profilstärke ist. Für größere Stufenanzahlen
genügt es in der Regel erneut, nur einen Typ von Paaren zu betrachten.
Die Arbeit schließt mit einer kurzen Diskussion und einem Ausblick in zukünftige
Forschungsfragen
- …
