The large number of real estate transactions across the United States, combined with closing process complexity, creates extremely large data sets that conceal anomalies indicative of fraud. The quantitative amount of damage due to fraud is immeasurable to the lives of individuals who are victims, not to mention the financial impact to organizations and the United States on a societal level. Based on data collected from a major title and settlement organization, the study creates a predictive model to detect transaction based anomalies in the residential market place. Using similar industry fraud detection models, such as credit card and automobile liability insurance, the study created a real estate domain specific model to detect fraud prior to closing the transaction. This statistical model utilizes data available during the transaction life cycle to determine potential risk and validates predictions using historical loss data from the organization. Near industry detection models were successfully leveraged to create a newly synthesized, quantitative detection model for the real estate problem domain. Organizations already have the power of detecting these potential fraud losses in the embedded knowledge base or data that exists within the organization itself. Data within an organization can be used in new ways to build competitive advantage while mitigating risk. This model provides foundational evidence to support the potential application of quantitative fraud detection within a transaction setting. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://www.proquest.com/en-US/products/dissertations/individuals.shtml.]