Linear programming (LP) stands for an optimization of a linear objective function, subject to linear and non-negativity constraints. For this purpose many methods for LP emerged. The best known is Simplex Method. Another group of methods for LP is represented by Interior Point Methods (IPM). These methods are based on interior points of feasible region of a problem, while Simplex Method uses basic feasible solution of a problem. This thesis focuses on theoretical background of IPM and brings it into relation with algorithms based on IPM. KKT system and its significance are included and the algorithm solving Linear Complementarity Problem is discussed as well. In this thesis, two algorithms based on IPM are introduced and used for solving a sample LP problem.

This thesis consists of four independent papers concerningdifferent aspects of interior methods for optimization. Threeof the papers focus on theoretical aspects while the fourth oneconcerns some computational experiments. The systems of equations solved within an interior methodapplied to a convex quadratic program can be viewed as weightedlinear least-squares problems. In the first paper, it is shownthat the sequence of solutions to such problems is uniformlybounded. Further, boundedness of the solution to weightedlinear least-squares problems for more general classes ofweight matrices than the one in the convex quadraticprogramming application are obtained as a byproduct. In many linesearch interior methods for nonconvex nonlinearprogramming, the iterates can "falsely" converge to theboundary of the region defined by the inequality constraints insuch a way that the search directions do not converge to zero,but the step lengths do. In the sec ond paper, it is shown thatthe multiplier search directions then diverge. Furthermore, thedirection of divergence is characterized in terms of thegradients of the equality constraints along with theasymptotically active inequality constraints. The third paper gives a modification of the analytic centerproblem for the set of optimal solutions in linear semidefiniteprogramming. Unlike the normal analytic center problem, thesolution of the modified problem is the limit point of thecentral path, without any strict complementarity assumption.For the strict complementarity case, the modified problem isshown to coincide with the normal analytic center problem,which is known to give a correct characterization of the limitpoint of the central path in that case. The final paper describes of some computational experimentsconcerning possibilities of reusing previous information whensolving system of equations arising in interior methods forlinear programming. <b>Keywords:</b>Interior method, primal-dual interior method,linear programming, quadratic programming, nonlinearprogramming, semidefinite programming, weighted least-squaresproblems, central path. <b>Mathematics Subject Classification (2000):</b>Primary90C51, 90C22, 65F20, 90C26, 90C05; Secondary 65K05, 90C20,90C25, 90C30.

<p>This thesis consists of four independent papers concerningdifferent aspects of interior methods for optimization. Threeof the papers focus on theoretical aspects while the fourth oneconcerns some computational experiments.</p><p>The systems of equations solved within an interior methodapplied to a convex quadratic program can be viewed as weightedlinear least-squares problems. In the first paper, it is shownthat the sequence of solutions to such problems is uniformlybounded. Further, boundedness of the solution to weightedlinear least-squares problems for more general classes ofweight matrices than the one in the convex quadraticprogramming application are obtained as a byproduct.</p><p>In many linesearch interior methods for nonconvex nonlinearprogramming, the iterates can "falsely" converge to theboundary of the region defined by the inequality constraints insuch a way that the search directions do not converge to zero,but the step lengths do. In the sec ond paper, it is shown thatthe multiplier search directions then diverge. Furthermore, thedirection of divergence is characterized in terms of thegradients of the equality constraints along with theasymptotically active inequality constraints.</p><p>The third paper gives a modification of the analytic centerproblem for the set of optimal solutions in linear semidefiniteprogramming. Unlike the normal analytic center problem, thesolution of the modified problem is the limit point of thecentral path, without any strict complementarity assumption.For the strict complementarity case, the modified problem isshown to coincide with the normal analytic center problem,which is known to give a correct characterization of the limitpoint of the central path in that case.</p><p>The final paper describes of some computational experimentsconcerning possibilities of reusing previous information whensolving system of equations arising in interior methods forlinear programming.</p><p><b>Keywords:</b>Interior method, primal-dual interior method,linear programming, quadratic programming, nonlinearprogramming, semidefinite programming, weighted least-squaresproblems, central path.</p><p><b>Mathematics Subject Classification (2000):</b>Primary90C51, 90C22, 65F20, 90C26, 90C05; Secondary 65K05, 90C20,90C25, 90C30.</p>

Cette thèse concerne quatre thèmes apparemment différents, mais en fait intimement liés : problèmes variationnels sur les algèbres de Jordan euclidiennes, problèmes de complémentarité sur l’espace des matrices symétriques, analyse angulaire entre deux cônes convexes fermés et analyse du chemin central en programmation conique symétrique.Dans la première partie de ce travail, le concept de “commutation au sens opérationnel” dans les algèbres de Jordan euclidiennes est étudié en fournissant un principe de commutation pour problèmes variationnels avec données spectrales.Dans la deuxième partie, nous abordons l’analyse et la résolution numérique d’une large classe de problèmes de complémentarité sur l’espace des matrices symétriques. Les conditions de complémentarité sont exprimées en termes de l’ordre de Loewner ou, plus généralement, en termes d’un cône du type Loewnerien.La troisième partie de ce travail est une tentative de construction d’une théorie générale des angles critiques pour une paire de cônes convexes fermés. L’analyse angulaire pour une paire de cônes spécialement structurés est également considérée. Par-exemple, nous travaillons avec des sous-espaces linéaires, des cônes polyédriques, des cônes de révolution, des cônes “topheavy” et des cônes de matrices.La dernière partie de ce travail étudie la convergence et le comportement asymptotique du chemin central en programmation conique symétrique. Ceci est fait en utilisant des techniques propres aux algèbres de Jordan. / This thesis deals with four different but interrelated topics: variational problems on Euclidean Jordan algebras, complementarity problems on the space of symmetric matrices, angular analysis between two closed convex cones and the central path for symmetric cone linear programming.In the first part of this work we study the concept of “operator commutation” in Euclidean Jordan algebras by providing a commutation principle for variational problems involving spectral data.Our main concern of the second part is the analysis and numerical resolution of a broad class of complementarity problems on spaces of symmetric matrices. The complementarity conditions are expressed in terms of the Loewner ordering or, more generally, with respect to a dual pair of Loewnerian cones.The third part of this work is an attempt to build a general theory of critical angles for a pair of closed convex cones. The angular analysis for a pair of specially structured cones is also covered. For instance, we work with linear subspaces, polyhedral cones, revolution cones, topheavy cones and cones of matrices.The last part of this work focuses on the convergence and the limiting behavior of the central path in symmetric cone linear programming. This is done by using Jordan-algebra techniques.

Studien syftar till att analysera equity- och debt baserad crowdfunding i förhållande till studiens primära teori om elaboration likelihood model (ELM) och hur dess variabler påverkar potentiella investerares investeringsbedömningar. Informationsasymmetri och behavioral finance har inkluderats i studien som två kompletterande delteorier. För att undersöka detta har studien avgränsats till potentiella investerare som geografiskt är bosatta i Stockholms län. Undersökningen har genomförts med hjälp av semistrukturerade intervjuer där totalt elva potentiella investerare har intervjuats. Respondenterna har baserats på ett snöbollsurval. Intervjuerna har skett genom att respondenterna tagit del av fyra scenarion som innehåller variabler kopplade till teorin om ELM. Studiens resultat påvisar att potentiella investerare bedömer samt väljer investeringsprojekt utifrån den centrala vägen i ELM teorin, som karaktäriseras av projektets kvalitet vilken är motsatsen till den perifera vägen som istället belyser elektronisk word of mouth (WoM). / The study aims to investigate how equity- and debt based crowdfunding in relation to the study’s primary theory of elaboration likelihood model (ELM) and its variables effect on potential investors’ investment decision. Information asymmetry and behavioral finance has also been included in the study as two complementary theories. To investigate this, the study has been delimited to potential investors who are geographically resident in Stockholm County. The survey has been conducted using semi-structured interviews, where a total of eleven potential investors have been interviewed based on a chain sampling. The potential investors have been presented four scenarios that contain variables linked to the theory of ELM. The study's findings show that potential investors assess and choose investment projects based on the central path of ELM theory, characterized by the quality of the project and opposite to the peripheral road that instead illuminates the electronic word of mouth (WoM).