2026-09-18T23:40:38Zhttps://dspace.mit.edu/server/oai/request

oai:dspace.mit.edu:1721.1/447932022-01-13T07:54:42Zcom_1721.1_7582com_1721.1_7581com_1721.1_141090com_1721.1_5427col_1721.1_131023col_1721.1_141481

Michael W. Golay. Holcombe, Robert (Robert Joseph) Massachusetts Institute of Technology. Dept. of Political Science. Massachusetts Institute of Technology. Department of Nuclear Science and Engineering Massachusetts Institute of Technology. Department of Political Science 2009-03-16T19:43:47Z 2009-03-16T19:43:47Z 2008 2008 http://hdl.handle.net/1721.1/44793 300312966 Thesis (S.M.)--Massachusetts Institute of Technology, Dept. of Nuclear Science and Engineering; and, (S.M.)--Massachusetts Institute of Technology, Dept. of Political Science, 2008. Includes bibliographical references (p. 110-112). Nuclear Proliferation is a complex problem that has plagued national security strategists since the advent of the first nuclear weapons. As the cost to produce nuclear weapons has continued to decline and the availability of nuclear material has become more widespread, the threat of proliferation has increased. The spread of technology and the globalization of the information age has made the threat not only more likely, but also more difficult to detect. Proliferation experts do not agree on the universal factors which cause nations to want to proliferate or the methods to prevent countries from successfully developing nuclear weapons. Historical evidence also indicates that the current nuclear powers pursued their nuclear programs for different reasons and under different conditions. This disparity presents a problem to decision makers who are tasked with preventing further nuclear proliferation. Bayesian Inference is a tool of quantitative analysis that is rapidly gaining interest in numerous fields of scientific study that have previously been limited to purely statistical methods. The Bayesian approach removes the statistical limitations of large-n data sets and strictly numerical types of data. It allows researchers to include sparse and rich data as well as qualitative data based on the opinions of subject matter experts. Bayesian inference allows the inclusion of both the quantitative data and subjective judgments in the determination of predictions about a theory of interest. This means that contrary to classic statistical methods, we can now make accurate predictions with reduced information and apply this probabilistic method to problems in social science. The problem of nuclear proliferation is one that lends itself to a Bayesian analysis. The data set is relatively small and the data is far from consistent from country to country. (cont.) There is however, a wide body of literature that seeks to explain proliferation factors and capabilities through both quantitative and qualitative means. This varied field can be brought together in a coherent method using Bayesian inference and specifically Bayesian Networks which graphically represent the various causal linkages. This work presents the development of a Bayesian Network describing the various causes, factors, and capabilities leading to proliferation. This network is constructed with conditional probabilities using theoretical insights and expert opinion. Bayesian inference using historical and real time events within the structure of the network is then used to give a decision maker an informed prediction of the proliferation danger of a specific country and inferences about which factors are causing it. by Robert Holcombe. S.M. 112 p. (some folded) eng Massachusetts Institute of Technology M.I.T. theses are protected by copyright. They may be viewed from this source for any purpose, but reproduction or distribution in any format is prohibited without written permission. See provided URL for inquiries about permission. http://dspace.mit.edu/handle/1721.1/7582 Nuclear Science and Engineering. Political Science. Development of a Bayesian Network to monitor the probability of nuclear proliferation Thesis application/pdf false Publication unknown unknown <Publication xmlns="https://www.openaire.eu/cerif-profile/1.1/" id="217f67ac-f052-4e89-92d8-4ca9b11480e2"> <Type xmlns="https://www.openaire.eu/cerif-profile/vocab/COAR_Publication_Types">http://purl.org/coar/resource_type/c_1843</Type> <Language>eng</Language> <Title>Development of a Bayesian Network to monitor the probability of nuclear proliferation</Title> <PublishedIn> <Publication> </Publication> </PublishedIn> <PublicationDate>2008</PublicationDate> <Authors> <Author> <DisplayName>Holcombe, Robert (Robert Joseph)</DisplayName> <Affiliation> <OrgUnit> </OrgUnit> </Affiliation> </Author> </Authors> <Editors> </Editors> <Publishers> <Publisher> <DisplayName>Massachusetts Institute of Technology</DisplayName> <OrgUnit /> </Publisher> </Publishers> <License>http://dspace.mit.edu/handle/1721.1/7582</License> <Keyword>Nuclear Science and Engineering.</Keyword> <Keyword>Political Science.</Keyword> <Abstract>Nuclear Proliferation is a complex problem that has plagued national security strategists since the advent of the first nuclear weapons. As the cost to produce nuclear weapons has continued to decline and the availability of nuclear material has become more widespread, the threat of proliferation has increased. The spread of technology and the globalization of the information age has made the threat not only more likely, but also more difficult to detect. Proliferation experts do not agree on the universal factors which cause nations to want to proliferate or the methods to prevent countries from successfully developing nuclear weapons. Historical evidence also indicates that the current nuclear powers pursued their nuclear programs for different reasons and under different conditions. This disparity presents a problem to decision makers who are tasked with preventing further nuclear proliferation. Bayesian Inference is a tool of quantitative analysis that is rapidly gaining interest in numerous fields of scientific study that have previously been limited to purely statistical methods. The Bayesian approach removes the statistical limitations of large-n data sets and strictly numerical types of data. It allows researchers to include sparse and rich data as well as qualitative data based on the opinions of subject matter experts. Bayesian inference allows the inclusion of both the quantitative data and subjective judgments in the determination of predictions about a theory of interest. This means that contrary to classic statistical methods, we can now make accurate predictions with reduced information and apply this probabilistic method to problems in social science. The problem of nuclear proliferation is one that lends itself to a Bayesian analysis. The data set is relatively small and the data is far from consistent from country to country.</Abstract> <Abstract>(cont.) There is however, a wide body of literature that seeks to explain proliferation factors and capabilities through both quantitative and qualitative means. This varied field can be brought together in a coherent method using Bayesian inference and specifically Bayesian Networks which graphically represent the various causal linkages. This work presents the development of a Bayesian Network describing the various causes, factors, and capabilities leading to proliferation. This network is constructed with conditional probabilities using theoretical insights and expert opinion. Bayesian inference using historical and real time events within the structure of the network is then used to give a decision maker an informed prediction of the proliferation danger of a specific country and inferences about which factors are causing it.</Abstract> <Access xmlns="http://purl.org/coar/access_right" > </Access> </Publication>