The release of Statistical Criteria for Selection of Functional Forms in Regression Analysis marks a significant contribution to the field of statistics and econometrics. Authored by Dr. K. Naga Vihari and Dr. M. Naresh, this scholarly work offers a comprehensive and methodical exploration of regression modeling, with particular emphasis on the critical task of selecting appropriate functional forms and regression models.
The book begins by laying a strong foundation in the theory of linear regression, clearly explaining classical assumptions, estimation techniques such as ordinary and restricted least squares, and the implications of violating these assumptions. To support advanced understanding, the authors systematically develop the necessary mathematical framework, including matrix algebra, Kronecker products, and system-based approaches to regression analysis.
A major strength of the book lies in its detailed treatment of model and regressor selection. The authors examine various statistical criteria based on mean square error, prediction efficiency, and widely used information criteria such as AIC, BIC, and related measures. These tools are presented with clarity and rigor, enabling readers to make informed and statistically sound modeling decisions.
Designed for researchers, postgraduate students, and professionals, this book serves as both a reference text and a practical guide, bridging theoretical depth with real-world applicability in regression analysis.