The skyline clause-also called the pareto clause-recently has been proposed as an extension to SQL it selects the ruples that are pareto optimal with respect to a set of designated skyline artribut es. This is the maximal vector problem in a relational context, but it represents a powerful extension to SQL which allows for the natural expression of on-line analytic processing (OLAP) queries and preferences in quires. Cardinality estimation of skyline sets is the focus in this work. A better understanding of skyline cardinality-and other perperties of the skyline-is useful for better design of skyline algorithms, is necessary to extend a query optimizer's cost model to accomodate skyline queries and helps to understand better to use skyline effectively for OLAP and preference queries.