View Complete Reference

Khosravani, A and Rasinariu, C (2012)

Transformation invariance of Benford variables and their numerical modeling

Recent Researches in Automatic Control and Electronics - Proceedings of the 14th International Conference on Automatic Control, Modelling & Simulation (ACMOS '12) and Proceedings of the 11th International Conference on Microelectronics, Nanoelectronics.

ISSN/ISBN: 978-1-61804-080-0 DOI: Not available at this time.



Abstract: A random variable is Benford distributed if the occurrence frequency of its most significant digit is P(d) = log10(1 + 1/d). Many empirical data sets obey this law with various degrees of accuracy .In this paper we analyze random variables X such that Y = 10X satisfies Benford's law exactly. We introduce a family of transformations on X that leave the digit distribution of Y invariant. Thus we identify new conditions for exact Benford conformance,and construct novel examples of such random variables. The Mathematica simulations strongly support our theoretical results.


Bibtex:
@InProceedings {, AUTHOR = {Khosravani, A. and Rasinariu, C.}, TITLE = {Transformation invariance of Benford variables and their numerical modeling}, BOOKTITLE = {Recent Researches in Automatic Control and Electronics - Proceedings of the 14th International Conference on Automatic Control, Modelling & Simulation (ACMOS '12) and Proceedings of the 11th International Conference on Microelectronics, Nanoelectronics}, YEAR = {2012}, ISBN = {978-1-61804-080-0}, PAGES = {57--61}, }


Reference Type: Conference Paper

Subject Area(s): Probability Theory, Statistics