Although many of the models commonly used in empirical finance are linear, the nature of financial data suggests that non-linear models are more appropriate for forecasting and accurately describing returns and volatility. The enormous number of non-linear time series models appropriate for modeling and forecasting economic time series models makes choosing the best model for a particular application daunting. This classroom-tested advanced undergraduate and graduate textbook, first published in 2000, provides a rigorous treatment of recently developed non-linear models, including regime-switching and artificial neural networks. The focus is on the potential applicability for describing and forecasting financial asset returns and their associated volatility. The models are analysed in detail and are not treated as 'black boxes'. Illustrated using a wide range of financial data, drawn from sources including the financial markets of Tokyo, London and Frankfurt.
Annotation An insightful and up-to-date study of the use of periodic models in the description and forecasting of economic data. Incorporating recent developments in the field, the authors investigate such areas as seasonal time series; periodic time series models; periodic integration; and periodic cointegration. The analysis from the inclusion of many new empirical examples and results.
With a new author team contributing decades of practical experience, this fully updated and thoroughly classroom-tested second edition textbook prepares students and practitioners to create effective forecasting models and master the techniques of time series analysis. Taking a practical and example-driven approach, this textbook summarises the most critical decisions, techniques and steps involved in creating forecasting models for business and economics. Students are led through the process with an entirely new set of carefully developed theoretical and practical exercises. Chapters examine the key features of economic time series, univariate time series analysis, trends, seasonality, aberrant observations, conditional heteroskedasticity and ARCH models, non-linearity and multivariate time series, making this a complete practical guide. Downloadable datasets are available online.
This book provides a self-contained account of periodic models for seasonally observed economic time series with stochastic trends. Two key concepts are periodic integration and periodic cointegration. Periodic integration implies that a seasonally varying differencing filter is required to remove a stochastic trend. Periodic cointegration amounts to allowing cointegration paort-term adjustment parameters to vary with the season. The emphasis is on useful econrameters and shometric models that explicitly describe seasonal variation and can reasonably be interpreted in terms of economic behaviour. The analysis considers econometric theory, Monte Carlo simulation, and forecasting, and it is illustrated with numerous empirical time series. A key feature of the proposed models is that changing seasonal fluctuations depend on the trend and business cycle fluctuations. In the case of such dependence, it is shown that seasonal adjustment leads to inappropriate results.
Advances in data collection and data storage techniques have enabled marketing researchers to study the individual characteristics of a large range of transactions and purchases, in particular the effects of household-specific characteristics. This 2001 book presents important and practically relevant quantitative models for marketing research. Each model is presented in detail with a self-contained discussion, which includes: a demonstration of the mechanics of the model, empirical analysis, real world examples, and interpretation of results and findings. The reader of the book will learn how to apply the techniques, as well as understand the methodological developments in the academic literature. Pathways are offered in the book for students and practitioners with differing numerical skill levels; a basic knowledge of elementary numerical techniques is assumed.