International Journal of Engineering Research and Development e-ISSN: 2278-067X, p-ISSN: 2278-800X, www.ijerd.com Volume 10, Issue 3 (March 2014), PP.64-71 64 Statistical Model For Risk Estimation 1 Dr. Vahida Attar, 2 Dr. D. Datta 1 Department of computer and IT, College of Engineering Pune ,Shivaji Nagar, Pune India. 2 Department of atomic energy, BRNS Mumbai, India. Abstract:- The study of the distribution and determinants of disease prevalence in man is done primarily in Radiation Epidemiology. Epidemiologists seek to relate risk of disease to different levels and patterns of radiation exposure. In this paper we examine statistical model of Poisson regression previously employed for the estimation of radiation risk. We examine different regression techniques, which overcome the underlying assumptions of Poisson Regression for risk estimation and propose Hurdle's Model for the same. The models need application of logarithmic transform to yield the additive model instead of multiplicative model, which is usually used in Risk Assessment and thus obtain the Linear Dose-Response Model. I. INTRODUCTION Epidemiology is concerned with study of distribution of disease and determinants of health-related states or events in specified human populations and application of this study for control of human health problems. It is observed that, people exposed to radiation usually suffer from cancer and other fatal diseases. For instance, there are two ways in which nuclear workers are exposed to radiation according to the Canadian Nuclear Safety Commission, either while working with sources of man-made radiation (nuclear industry, health care, research institutions or manufacturing) or they are exposed to elevated levels of natural radiation (mining, air crews construction). Radiation is categorized as ionizig and non-ionizing.[8] Ionizing radiation is radiation with enough energy so that during an interaction with an atom it can remove bound electrons, i.e., it can ionize atoms. Examples are X-Rays and electrons.Non-ionizing radiation is radiation without enough energy to remove bound electrons from their orbits around atoms. Examples are microwaves and visible light.The ionizing radiation interacts with the cells and damages them, which in turn results in malignant growth in the body. Thus studying the levels of radiation and its corresponding effects can prove beneficial in setting the safety levels of exposure. In our study we intend to develop a generalized model for risk evaluation or odds ratio for death due to cardiovascular disease and other cancer due to ionizing radiation. Radiation Effects Research Foundation has played an important role of carrying out cohort study on the Japanese Atomic Bomb survivors with a follow-up of 50 years. This report makes use of data obtained from the Radiation Effects Research Foundation (RERF), Hiroshima and Nagasaki, Japan. RERF is a private, non-profit foundation funded by the Japanese Ministry of Health, Labour and Welfare (MHLW) and the U.S. Department of Energy, the latter through the National Academy of Sciences. The objective is to apply suitable methods to gain further insight in the model developed by RERF on Cardiovascular diseases. Main outcome is to measure Mortality from stroke or heart disease as the underlying cause of death and dose response relations with atomic bomb radiation. This finding would significantly benefit the humanity as with growing technology, there are associated hazards. This is an interdisciplinary problem as it encompasses Epidemiology, Statistics and Data Analysis Methods. The paper starts discussion of prevalent risk models for low-ionized radiation. Aanalyse the strengths and limitations of the models used. This is followed by provision of theoretical background of Poisson Model, used for count data and estimation of relative risk. It comprises of multiplicative and additive models of relative risk. Finally it proposes the use of variations of Poisson Regression called Hurdle model. II. RISK OF RADIATION EXPOSURE Comparing the acute exposure experienced by atomic bomb survivors with the low dose rate exposures experienced gradually over time due to occupational, environmental or natural background circumstances is now frequently done. Wide ranges of risk estimates have been reported with some significantly lower than the