Portfolio selection has a long tradition in financial economics and plays an integral role in investment management. Portfolio selection provides the framework to determine optimal portfolio choice from a universe of available investments. However, the asset weightings from portfolio selection are optimal only if the empirical characteristics of asset returns do not violate the portfolio selection model assumptions. This thesis explores the empirical characteristics of traditional assets and hedge fund returns and examines their effects on the assumptions of linearity-in-the-mean testing and portfolio selection. The encompassing theme of this thesis is the empirical interplay between traditional assets and hedge fund returns. Despite the paucity of hedge fund research, pension funds continue to increase their portfolio allocations to global hedge funds in an effort to pursue higher risk-adjusted returns. This thesis presents three empirical studies which provide positive insights into the relationships between traditional assets and hedge fund returns. The first two empirical studies examine an emerging body of literature which suggests that the relationship between traditional assets and hedge fund returns is non-linear. For mean-variance investors, non-linear asset returns are problematic as they do not satisfy the assumption of linearity required for the covariance matrix in portfolio selection. To examine the linearity assumption as it relates to a mean-variance investor, a hypothesis test approach is employed which investigates the linearity-in-the-mean of traditional assets and hedge funds. The findings from the first two empirical studies reveal that conventional linearity-in-the-mean tests incorrectly conclude that asset returns are nonlinear. We demonstrate that the empirical characteristics of heteroscedasticity and autocorrelation in asset returns are the primary sources of test mis-specification in these linearity-in-the-mean hypothesis tests. To address this problem, an innovative approach is proposed to control heteroscedasticity and autocorrelation in the underlying tests and it is shown that traditional assets and hedge funds are indeed linear-in-the-mean. The third and final study of this thesis explores traditional assets and hedge funds in a portfolio selection framework. Following the theme of the previous two studies, the effects of heteroscedasticity and autocorrelation are examined in the portfolio selection context. The characteristics of serial correlation in bond and hedge fund returns are shown to cause a downward bias in the second sample moment. This thesis proposes two methods to control for this effect and it is shown that autocorrelation induces an overallocation to bonds and hedge funds. Whilst heteroscedasticity cannot be directly examined in portfolio selection, empirical evidence suggests that heteroscedastic events (such as those that occurred in August 1998) translate into the empirical feature known as tail-risk. The effects of tail-risk are examined by comparing the portfolio decisions of mean-variance analysis (MVA) versus mean-conditional value at risk (M-CVaR) investors. The findings reveal that the volatility of returns in a MVA portfolio decreases when hedge funds are included in the investment opportunity set. However, the reduction in the volatility of portfolio returns comes at a cost of undesirable third and fourth moments. Furthermore, it is shown that investors with M-CVaR preferences exhibit a decreasing demand for hedge funds as their aversion for tail-risk increases. The results of the thesis highlight the sensitivities of linearity tests and portfolio selection to the empirical features of heteroscedasticity, autocorrelation and tail-risk. This thesis contributes to the literature by providing refinements to these frameworks which allow improved inferences to be made when hedge funds are examined in linearity and portfolio selection settings.

Parmi les formes d’organisation de l’habitat antique, il en est une qui est longtemps restée en retrait dans la recherche archéologique française. Les agglomérations antiques, qui correspondent à des formes très variées d’habitat groupé, suscitent un intérêt majeur depuis une vingtaine d’années.L’étude historiographique, à l’échelle de la Gaule, fait ressortir un espace en marge des recherches, celui des cités du Massif central. Cette thèse propose donc d’étudier les agglomérations antiques des cités arverne, vellave, gabale, rutène, cadurque et lémovice entre le Ier s. av. J.-C. et le Ve s. ap. J.-C. Les objectifs sont multiples. Il a été nécessaire dans un premier temps d’établir un corpus de sites à partir de la bibliographie régionale. Celui-ci est composé de notices normalisées au sein desquelles l’ensemble des données actuellement disponibles a été présenté. Pour répondre aux questions d’organisation spatiale et de classification, les données bibliographiques recueillies, complétées par des opérations de terrain (prospections terrestres, aériennes, géophysiques, relevés topographiques), ont été synthétisées à partir d’une série de descripteurs archéologiques développés sur le modèle d’Archaeomedes. L’emploi de ces descripteurs a permis la mise en œuvre d’analyses statistiques multivariées afin de dépasser le simple classement par niveaux de fiabilité (sites rejetés, agglomérations hypothétiques, agglomérations avérées). C’est sur la base de la classification retenue après différents tests et sur la hiérarchie urbaine proposée que les analyses spatiales et morphologiques permettent de brosser un tableau complet de l’armature urbaine des cités antiques du Massif central.Les résultats obtenus ont permis d’exposer et de comprendre à la fois les relations entre les agglomérations et le milieu naturel mais aussi avec leur environnement archéologique par l’étude de leurs relations avec l’habitat rural, les chefs-lieux, les limites des cités, leur territoire théorique, les axes de communications terrestres et fluviaux. En changeant d’échelle, l’analyse interne des agglomérations permet d’appréhender à la fois leur morphologie générale mais aussi la place des édifices monumentaux au sein des agglomérations et au sein des cités, le rôle de l’artisanat, les aménagements hydrauliques, … Pour l’analyse de l’armature urbaine, la thèse prend nécessairement en compte l’évolution chronologique des agglomérations à travers leur origine protohistorique, leur évolution durant l’Antiquité et leur devenir au début du Moyen Âge.Pour mener à bien ce travail, il a été nécessaire de croiser les approches documentaires (bibliographie, opérations de terrains), de croiser les échelles d’analyse (du site au Massif central) et de comparer les résultats à ceux des régions voisines. Passées, les analyses statistiques, c’est à partir d’un SIG et d’une base de données que l’ensemble des synthèses a pu être mené.Cette thèse constitue la première synthèse sur les agglomérations antiques des cités du Massif central. Les résultats mettent en avant l’importance des agglomérations dans le système de peuplement et des armatures urbaines variables entre les cités mais qui semblent structurées en partie par le relief et les axes de communications. Les fonctions présentes au sein des agglomérations, si elles sont variées et récurrentes, montrent des spécificités entre les cités mais aussi une hétérogénéité au sein de chacune d’elles. Enfin, l’analyse chronologique montre une réduction du nombre des agglomérations à partir du IIIe siècle mais pas leur disparition. De même, de nouvelles formes d’habitats groupés apparaissent à partir du IVe siècle attestant non pas une crise mais une évolution dans l’organisation urbaine des cités pour répondre à de nouveaux besoins. / Among the forms of organization one can observe in the ancient habitat, there is one particular structure that has been slightly ignored by French archeologists: the lesser urbanized settlements. However, in the last twenty years, small towns have attracted renewed interest.A historiographical study of the Gaul territory highlighted a less studied group of population: the ancient peoples of the Massif Central. In this thesis we intend to have a closer look at such populations as the Arvernes, the Vellaves, the Gabales, the Rutenes, the Cadurques, and the Lemovices from 1st century B.C. - 5th century A.D. We first established a corpus of sites starting with the study of the regional bibliography, which enabled us to create records for each area. We completed this first approach with several field operations such as ground, aerial, geophysical and topographical surveys. All those data were synthesized in a series of archaeological descriptors developed using the Archaeomedes model.The use of such descriptors enabled us to develop some statistical analyses and go beyond the simple reliability factor (which allows the distinction between rejected sites, hypothetical and true urbanized clusters). On the basis of our tested and proven classification, keeping in mind the suggested urban hierarchy and with the help of our spatial and morphological analyses, we were able to obtain a more complete picture of the urban framework of the ancient cities of the Massif Central.The data collected allowed us to better apprehend and understand the relation between such urban clusters and their natural environment, but also to place them in an archaeological context by studying the links between rural housing, the ancient administrative structure (chef-lieu), the city and territory limits, and the major communication routes (land and water). Through a multi-scale study of those population clusters, we were able to determine their general structure, to look at the role of local crafts and trade but also the place of major buildings, and waterworks. We took specific care to the chronological evolution in thosesmall towns by looking at them from their protohistoric beginnings through Antiquity and until early medieval times.While we cross-referenced all the different sources (bibliography and field surveys) and the analysis scales within the Massif Central, we also did a cross-comparison with other neighboring areas in order to validate such work. Besides the statistical data, we used the Geographic Information System and our database to reach our conclusions.This thesis aims at uncovering the reality of those ancient population clusters present in the Massif Central. Our results showed clearly the importance of those in the settlement system, their different forms structured according to the local geography and major trade routes. The functions/positions available in small towns were varied and recurrent but also corresponded to specific trades within each structure while remaining homogeneous.Through our chronological analysis we were able to point out a significant reduction in their numbers starting in the third century, though not their disappearing. In the 4th century new types of population clusters appeared, not as a response to an acute crisis, but more as a result of an evolution called on by new needs.