Time series play a crucial role in modern economies at all levels of activity and are used by decision makers to plan for a better future. Before publication time series are subject to statistical adjustments and this is the first statistical book to systematically deal with the methods most often applied for such adjustments. Regression-based models are emphasized because of their clarity, ease of application, and superior results. Each topic is illustrated with real case examples. In order to facilitate understanding of their properties and limitations of the methods discussed a real data example is followed throughout the book.

The use of computational methods in statistics to face complex problems and highly dimensional data, as well as the widespread availability of computer technology, is no news. The range of applications, instead, is unprecedented. As often occurs, new and complex data types require new strategies, demanding for the development of novel statistical methods and suggesting stimulating mathematical problems. This book is addressed to researchers working at the forefront of the statistical analysis of complex systems and using computationally intensive statistical methods.

Statistical offices have often recourse to benchmarking methods for compiling quarterly national accounts (QNA). Benchmarking methods employ quarterly indicator series (i) to distribute annual, more reliable series of national accounts and (ii) to extrapolate the most recent quarters not yet covered by annual benchmarks. The Proportional First Differences (PFD) benchmarking method proposed by Denton (1971) is a widely used solution for distribution, but in extrapolation it may suffer when the movements in the indicator series do not match consistently the movements in the target annual benchmarks. For this reason, an enhanced formula for extrapolation was recommended by the IMF’s Quarterly National Accounts Manual: Concepts, Data Sources, and Compilation (2001). We discuss the rationale behind this technique, and propose a matrix formulation of it. In addition, we present applications of the enhanced formula to artificial and real-life benchmarking examples showing how the extrapolations for the most recent quarters can be improved.

This volume includes contributions selected after a double blind review process and presented as a preliminary version at the 45th Meeting of the Italian Statistical Society. The papers provide significant and innovative original contributions and cover a broad range of topics including: statistical theory; methods for time series and spatial data; statistical modeling and data analysis; survey methodology and official statistics; analysis of social, demographic and health data; and economic statistics and econometrics.

This revised book presents theoretical results relevant to Edgeworth and saddlepoint expansions to densities and distribution functions. It provides examples of their application in some simple and a few complicated settings, along with numerical, as well as asymptotic, assessments of their accuracy. Variants on these expansions, including much of modern likelihood theory, are discussed and applications to lattice distributions are extensively treated.

Applied Data Mining for Forecasting Using SAS, by Tim Rey, Arthur Kordon, and Chip Wells, introduces and describes approaches for mining large time series data sets. Written for forecasting practitioners, engineers, statisticians, and economists, the book details how to select useful candidate input variables for time series regression models in environments when the number of candidates is large, and identifies the correlation structure between selected candidate inputs and the forecast variable. This book is essential for forecasting practitioners who need to understand the practical issues involved in applied forecasting in a business setting. Through numerous real-world examples, the authors demonstrate how to effectively use SAS software to meet their industrial forecasting needs. This book is part of the SAS Press program.

This book offers a new, fairly efficient, and robust alternative to analyzing multivariate data. The analysis of data based on multivariate spatial signs and ranks proceeds very much as does a traditional multivariate analysis relying on the assumption of multivariate normality; the regular L2 norm is just replaced by different L1 norms, observation vectors are replaced by spatial signs and ranks, and so on. A unified methodology starting with the simple one-sample multivariate location problem and proceeding to the general multivariate multiple linear regression case is presented. Companion estimates and tests for scatter matrices are considered as well. The R package MNM is available for computation of the procedures. This monograph provides an up-to-date overview of the theory of multivariate nonparametric methods based on spatial signs and ranks. The classical book by Puri and Sen (1971) uses marginal signs and ranks and different type of L1 norm. The book may serve as a textbook and a general reference for the latest developments in the area. Readers are assumed to have a good knowledge of basic statistical theory as well as matrix theory. Hannu Oja is an academy professor and a professor in biometry in the University of Tampere. He has authored and coauthored numerous research articles in multivariate nonparametrical and robust methods as well as in biostatistics.