Believing in a single model may be dangerous, and addressing model uncertainty by averaging different models in making forecasts may be very beneficial. In this thesis we focus on forecasting financial time series using model averaging schemes as a way to produce optimal forecasts. We derive and discuss in simulation exercises and empirical applications model averaging techniques that can reproduce stylized facts of financial time series, such as low predictability and time-varying patterns. We emphasize that model averaging is not a "magic" methodology which solves a priori problems of poorly forecasting. Averaging techniques have an essential requirement: individual models have to fit data. In the first section we provide a general outline of the thesis and its contributions to previ ous research. In Chapter 2 we focus on the use of time varying model weight combinations. In Chapter 3, we extend the analysis in the previous chapter to a new Bayesian averaging scheme that models structural instability carefully. In Chapter 4 we focus on forecasting the term structure of U.S. interest rates. In Chapter 5 we attempt to shed more light on forecasting performance of stochastic day-ahead price models. We examine six stochastic price models to forecast day-ahead prices of the two most active power exchanges in the world: the Nordic Power Exchange and the Amsterdam Power Exchange. Three of these forecasting models include weather forecasts. To sum up, the research finds an increase of forecasting power of financial time series when parameter uncertainty, model uncertainty and optimal decision making are included.

Complex engineering systems such as bridges, roads, flood defence structures, and power pylons play an important role in our society. Unfortunately such systems are subject to deterioration, meaning that in course of time their condition falls from higher to lower, and possibly even to unacceptable, levels. Maintenance actions such as inspection, local repair and replacement should be done to retain such systems in or restore them to acceptable operating conditions. After all, the economic consequences of malfunctioning infrastructure systems can be huge. In the life-cycle management of engineering systems, the decisions regarding the timing and the type of maintenance depend on the temporal uncertainty associated with the deterioration. Hence it is of importance to model this uncertainty. In the literature, deterioration models based on Brownian motion and gamma process have had much attention, but a thorough comparison of these models lacks. In this thesis both models are compared on several aspects, both in a theoretical as well as in an empirical setting. Moreover, they are compared with physical process models, which can capture structural insights into the underlying process. For the latter a new framework is developed to draw inference. Next, models for imperfect maintenance are investigated. Finally, a review is given for systems consisting of multiple components.

A complete set of statistical tools for beginning financial analysts from a leading authority Written by one of the leading experts on the topic, An Introduction to Analysis of Financial Data with R explores basic concepts of visualization of financial data. Through a fundamental balance between theory and applications, the book supplies readers with an accessible approach to financial econometric models and their applications to real-world empirical research. The author supplies a hands-on introduction to the analysis of financial data using the freely available R software package and case studies to illustrate actual implementations of the discussed methods. The book begins with the basics of financial data, discussing their summary statistics and related visualization methods. Subsequent chapters explore basic time series analysis and simple econometric models for business, finance, and economics as well as related topics including: Linear time series analysis, with coverage of exponential smoothing for forecasting and methods for model comparison Different approaches to calculating asset volatility and various volatility models High-frequency financial data and simple models for price changes, trading intensity, and realized volatility Quantitative methods for risk management, including value at risk and conditional value at risk Econometric and statistical methods for risk assessment based on extreme value theory and quantile regression Throughout the book, the visual nature of the topic is showcased through graphical representations in R, and two detailed case studies demonstrate the relevance of statistics in finance. A related website features additional data sets and R scripts so readers can create their own simulations and test their comprehension of the presented techniques. An Introduction to Analysis of Financial Data with R is an excellent book for introductory courses on time series and business statistics at the upper-undergraduate and graduate level. The book is also an excellent resource for researchers and practitioners in the fields of business, finance, and economics who would like to enhance their understanding of financial data and today's financial markets.

Monitoring involves the collection, analysis and evaluation of information over time. For many professionals, monitoring is a central aspect of their work. For example, policy- makers closely watch the e®ects of their current policies to set the right course for reform. Likewise, physicians monitor the well-being of their patients to adjust their treatments when necessary. In business, n̄ancial investors monitor stock prices and interest rates to optimally time their investments, while marketing managers watch their customers' needs and wants to frame their marketing e®orts. The above examples illustrate that monitoring is crucial in many disciplines to make the right decisions at the right moment. For this reason, there has always been a need for improved monitoring methods. With the advent of increasingly powerful computers and advanced analytical techniques, monitoring systems can nowadays process large amounts of information and have become fully automated where desired. A large body of moni- toring methods originate from academics. Especially during the past four decades, many insights from various ēlds such as economics, statistics, psychometrics and econometrics found their way into everyday monitoring practice. With the overwhelming availability of information in some cases, but also the intrinsic lack of information in other cases, the area is continuously faced with new and highly relevant research challenges. The aim of this thesis is to contribute to the development of new monitoring methods by o®ering potential solutions to some of these challenges. The challenges studied in this thesis arise from all three aspects of monitoring, that is from the collection, the analysis as well as from the evaluation of information.