Publisher's Synopsis
Knowledge–based business decisions and economic policy decisions are needed in an open competitive business and economic environment. This calls for the search for the best possible econometric model to build the best possible knowledge–base. This book offers the necessary tools for such knowledge–based decisions.
This book provides an integrated presentation of statistical theory, methods and computation algorithm, enabling the researcher to make suitable modifications and improvements to suit his or her purpose. A variety of illustrative models, and model selection and estimation methods offer a wide menu from which the reader can select the most appropriate and best–fitting model in any practical situation.
The topics covered look at how to choose independent variables, the functional forms in which they should appear, the choice between alternate models, best estimation of parameters, and testing of statistical hypotheses. The problem and data setting considered are time series, cross–section, time–series of cross–sections, multivariate and univariate models, single equation and system of equations models, models with unobservable variables, models with limited dependent variables, and models with mixtures of statistical distributions.
This book:
- Looks at new methods including nonparametric and semi–parametric regression, partial least squares, robust regression, path analysis, canonical correlation, bootstrap and evolutionary algorithm.
- Provides user–friendly computing software.
- Draws heavily from the teaching and research experience of all the authors in China, India, and the United States.
- Statistical theories and methods are illustrated throughout.
The book is useful for graduate students in economics and business, and for researchers and consultants involved in business analytics.
The research work of this book was supported by National Science Foundation of China.
