Statistical Estimation Of Optimal Portfolios For Dependent Returns Of Assets

The composition of portfolios is one of the most fundamental and important methods in financial engineering, used to control the risk of investments. This book provides a comprehensive overview of statistical inference for portfolios and their various applications. A variety of asset processes are introduced, including non-Gaussian stationary processes, nonlinear processes, non-stationary processes, and the book provides a framework for statistical inference using local asymptotic normality (LAN). The approach is generalized for portfolio estimation, so that many important problems can be covered. This book can primarily be used as a reference by researchers from statistics, mathematics, finance, econometrics, and genomics. It can also be used as a textbook by senior undergraduate and graduate students in these fields.

We address three key aspects of optimal portfolio construction: expected return, variance-covariance modeling and optimization in presence of cardinality constraints. On expected return modeling, we extend the self-excited point process framework to model conditional arrival intensities of bid and ask side market orders of listed stocks. The cross-excitation of market orders is modeled explicitly such that the ask side market order size and bid side probability weighted order book cumulative volume can affect the ask side order intensity, and vice versa. Different variations of the framework are estimated by using method of maximum likelihood estimation, based on a recursive application of the log-likelihood functions derived in this thesis. Results indicate that the self-excited point process framework is able to capture a significant amount of the underlying trading dynamics of market orders, both in-sample and out-of-sample. A new framework is introduced, Realized GARCH, for the joint modeling of returns and realized measures of volatility. A key feature is a measurement equation that relates the realized measure to the conditional variance of returns. The measurement equation facilitates a simple modeling of the dependence between returns and future volatility. Realized GARCH models with a linear or log-linear specification have many attractive features. They are parsimonious, simple to estimate, and imply an ARMA structure for the conditional variance and the realized measure. An empirical application with DJIA stocks and an exchange traded index fund shows that a simple Realized GARCH structure leads to substantial improvements in the empirical fit over standard GARCH models. Finally we describe a novel algorithm to obtain the solution of the optimal portfolio problem with NP-hard cardinality constraints. The algorithm is based on a local relaxation that exploits the inherent structure of the objective function. It solves a sequence of small, local, quadratic-programs by first projecting asset returns onto a reduced metric space, followed by clustering in this space to identify sub-groups of assets that best accentuate a suitable measure of similarity amongst different assets. The algorithm can either be cold started using the centroids of initial clusters or be warm started based on the output of a previous result. Empirical result, using baskets of up to 3,000 stocks and with different cardinality constraints, indicates that the algorithm is able to achieve significant performance gain over a sophisticated branch-and-cut method. One key application of this local relaxation algorithm is in dealing with large scale cardinality constrained portfolio optimization under tight time constraint, such as for the purpose of index tracking or index arbitrage at high frequency.

The composition of portfolios is one of the most fundamental and important methods in financial engineering, used to control the risk of investments. This book provides a comprehensive overview of statistical inference for portfolios and their various applications. A variety of asset processes are introduced, including non-Gaussian stationary processes, nonlinear processes, non-stationary processes, and the book provides a framework for statistical inference using local asymptotic normality (LAN). The approach is generalized for portfolio estimation, so that many important problems can be covered. This book can primarily be used as a reference by researchers from statistics, mathematics, finance, econometrics, and genomics. It can also be used as a textbook by senior undergraduate and graduate students in these fields.

Since the formalization of asset allocation in 1952 with the publication of Portfolio Selection by Harry Markowitz, there have been great strides made to enhance the application of this groundbreaking theory. However, progress has been uneven. It has been punctuated with instances of misleading research, which has contributed to the stubborn persistence of certain fallacies about asset allocation. A Practitioner's Guide to Asset Allocation fills a void in the literature by offering a hands-on resource that describes the many important innovations that address key challenges to asset allocation and dispels common fallacies about asset allocation. The authors cover the fundamentals of asset allocation, including a discussion of the attributes that qualify a group of securities as an asset class and a detailed description of the conventional application of mean-variance analysis to asset allocation.. The authors review a number of common fallacies about asset allocation and dispel these misconceptions with logic or hard evidence. The fallacies debunked include such notions as: asset allocation determines more than 90% of investment performance; time diversifies risk; optimization is hypersensitive to estimation error; factors provide greater diversification than assets and are more effective at reducing noise; and that equally weighted portfolios perform more reliably out of sample than optimized portfolios. A Practitioner's Guide to Asset Allocation also explores the innovations that address key challenges to asset allocation and presents an alternative optimization procedure to address the idea that some investors have complex preferences and returns may not be elliptically distributed. Among the challenges highlighted, the authors explain how to overcome inefficiencies that result from constraints by expanding the optimization objective function to incorporate absolute and relative goals simultaneously. The text also explores the challenge of currency risk, describes how to use shadow assets and liabilities to unify liquidity with expected return and risk, and shows how to evaluate alternative asset mixes by assessing exposure to loss throughout the investment horizon based on regime-dependent risk. This practical text contains an illustrative example of asset allocation which is used to demonstrate the impact of the innovations described throughout the book. In addition, the book includes supplemental material that summarizes the key takeaways and includes information on relevant statistical and theoretical concepts, as well as a comprehensive glossary of terms.

We address three key aspects of optimal portfolio construction: expected return, variance-covariance modeling and optimization in presence of cardinality constraints. On expected return modeling, we extend the self-excited point process framework to model conditional arrival intensities of bid and ask side market orders of listed stocks. The cross-excitation of market orders is modeled explicitly such that the ask side market order size and bid side probability weighted order book cumulative volume can affect the ask side order intensity, and vice versa. Different variations of the framework are estimated by using method of maximum likelihood estimation, based on a recursive application of the log-likelihood functions derived in this thesis. Results indicate that the self-excited point process framework is able to capture a significant amount of the underlying trading dynamics of market orders, both in-sample and out-of-sample. A new framework is introduced, Realized GARCH, for the joint modeling of returns and realized measures of volatility. A key feature is a measurement equation that relates the realized measure to the conditional variance of returns. The measurement equation facilitates a simple modeling of the dependence between returns and future volatility. Realized GARCH models with a linear or log-linear specification have many attractive features. They are parsimonious, simple to estimate, and imply an ARMA structure for the conditional variance and the realized measure. An empirical application with DJIA stocks and an exchange traded index fund shows that a simple Realized GARCH structure leads to substantial improvements in the empirical fit over standard GARCH models. Finally we describe a novel algorithm to obtain the solution of the optimal portfolio problem with NP-hard cardinality constraints. The algorithm is based on a local relaxation that exploits the inherent structure of the objective function. It solves a sequence of small, local, quadratic-programs by first projecting asset returns onto a reduced metric space, followed by clustering in this space to identify sub-groups of assets that best accentuate a suitable measure of similarity amongst different assets. The algorithm can either be cold started using the centroids of initial clusters or be warm started based on the output of a previous result. Empirical result, using baskets of up to 3,000 stocks and with different cardinality constraints, indicates that the algorithm is able to achieve significant performance gain over a sophisticated branch-and-cut method. One key application of this local relaxation algorithm is in dealing with large scale cardinality constrained portfolio optimization under tight time constraint, such as for the purpose of index tracking or index arbitrage at high frequency.

Until now, few systematic studies of optimal statistical inference for stochastic processes had existed in the financial engineering literature, even though this idea is fundamental to the field. Balancing statistical theory with data analysis, Optimal Statistical Inference in Financial Engineering examines how stochastic models can effectively des

While mainstream financial theories and applications assume that asset returns are normally distributed, overwhelming empirical evidence shows otherwise. Yet many professionals don’t appreciate the highly statistical models that take this empirical evidence into consideration. Fat-Tailed and Skewed Asset Return Distributions examines this dilemma and offers readers a less technical look at how portfolio selection, risk management, and option pricing modeling should and can be undertaken when the assumption of a non-normal distribution for asset returns is violated. Topics covered in this comprehensive book include an extensive discussion of probability distributions, estimating probability distributions, portfolio selection, alternative risk measures, and much more. Fat-Tailed and Skewed Asset Return Distributions provides a bridge between the highly technical theory of statistical distributional analysis, stochastic processes, and econometrics of financial returns and real-world risk management and investments.

This book compiles theoretical developments on statistical inference for time series and related models in honor of Masanobu Taniguchi's 70th birthday. It covers models such as long-range dependence models, nonlinear conditionally heteroscedastic time series, locally stationary processes, integer-valued time series, Lévy Processes, complex-valued time series, categorical time series, exclusive topic models, and copula models. Many cutting-edge methods such as empirical likelihood methods, quantile regression, portmanteau tests, rank-based inference, change-point detection, testing for the goodness-of-fit, higher-order asymptotic expansion, minimum contrast estimation, optimal transportation, and topological methods are proposed, considered, or applied to complex data based on the statistical inference for stochastic processes. The performances of these methods are illustrated by a variety of data analyses. This collection of original papers provides the reader with comprehensive and state-of-the-art theoretical works on time series and related models. It contains deep and profound treatments of the asymptotic theory of statistical inference. In addition, many specialized methodologies based on the asymptotic theory are presented in a simple way for a wide variety of statistical models. This Festschrift finds its core audiences in statistics, signal processing, and econometrics.

An introduction to the theory and methods of empirical asset pricing, integrating classical foundations with recent developments. This book offers a comprehensive advanced introduction to asset pricing, the study of models for the prices and returns of various securities. The focus is empirical, emphasizing how the models relate to the data. The book offers a uniquely integrated treatment, combining classical foundations with more recent developments in the literature and relating some of the material to applications in investment management. It covers the theory of empirical asset pricing, the main empirical methods, and a range of applied topics. The book introduces the theory of empirical asset pricing through three main paradigms: mean variance analysis, stochastic discount factors, and beta pricing models. It describes empirical methods, beginning with the generalized method of moments (GMM) and viewing other methods as special cases of GMM; offers a comprehensive review of fund performance evaluation; and presents selected applied topics, including a substantial chapter on predictability in asset markets that covers predicting the level of returns, volatility and higher moments, and predicting cross-sectional differences in returns. Other chapters cover production-based asset pricing, long-run risk models, the Campbell-Shiller approximation, the debate on covariance versus characteristics, and the relation of volatility to the cross-section of stock returns. An extensive reference section captures the current state of the field. The book is intended for use by graduate students in finance and economics; it can also serve as a reference for professionals.