Abstract not found

Abstract not found

Abstract not found

All in-text references underlined in blue are linked to publications on ResearchGate, letting you access and read them immediately.

We develop new tail-trimmed M-estimation methods for heavy tailed Nonlinear AR-GARCH models. Tail-trimming allows both identi…cation of the true parameter and asymptotic normality for nonlinear models with asymmetric errors. In heavy tailed cases the rate of convergence is in…nitesimally close to the highest possible amongst M-estimators for a particular loss function, hence super- 12-convergence can be achieved in nonlinear AR models with in…nite variance errors, and arbitrarily near 12-convergence for GARCH with errors that have an in…nite fourth moment. We present a consistent estimator of the covariance matrix that permits classic inference without knowledge of the rate of convergence, and explore asymptotic covariance and bootstrap mean-squared-error methods for selecting trimming parameters. A simulation study shows the estimator trumps existing ones for AR and GARCH models based on sharpness, approximate normality, rate of convergence, and test accuracy. We then use the estimator to study asset returns data. 1. INTRODUCTION It is now widely accepted that log-returns of many macroeconomic and …nancial time series are heavy tailed, asymmetrically distributed, and exhibit clustering of large values. In broader contexts extremes are encountered in actuarial, meteorological, and

in these networks, and (3) the extremely limited computational and energy resources available. Until now, the burden of handling these issues was put on the software application developer. This dissertation presents three novel programming models and middleware systems that address these challenges: Limone, Agilla

of scalability and maintainability. Hence, different approaches are based on Peer-to-Peer (P2P) technology. In particular, fast-paced NVEs stress these systems and most likely cause high overlay maintenance overhead. This Master’s Thesis presents a novel approach to spatial Publish/Subscribe (Pub/Sub) in P2P

April 29, 2011Für UnsShe had never before seen a rabbit with either a waistcoat-pocket, or a watch to take out of it, and burning with curiosity, she ran across the field after it Lewis CarrollThe following document, written under the supervision of Dr. reviewed by:

I would like to thank Prof. Dr. Torsten Braun, head of the Computer Network and Distributed Systems group (RVS), for supervising this work and for his insightful advises. Prof. Dr. Torsten Braun encouraged and motivated me to publish my research results and he provided me the opportunity to present

are client-server based, which is feasible for relatively slow role-playing games. Those have modest bandwidth and latency requirements and are paid for by their customers. For MMOGs with higher realtime requirements and/or a smaller number of customers willing to pay, peer-to-peer networking seems to be a