Abstract: Various embodiments disclosed include, for each entity in a portfolio, receiving entity data indicative of attributes of an entity, determining the received entity data for at least some entities is missing a portion of the entity data required to perform a cyber risk analysis; and synthesizing the missing portion. The method may further include comparing the received entity data and synthesized missing portion for each of the entities to each other; locating clusters of similar entity data shared between two or more of the entities; and calculating a cyber risk score representing how different the entities are to one another based on the entity data that are not shared between entities. Some embodiments include comparing entities that are missing some entity data to entities which have complete entity data, and generating a synthesized portfolio by selecting entities having complete entity data to replace the entities that are missing entity data.