This monograph addresses two quite different topics, in the belief that each can shed light on the other. Firstly, it lays the foundation for a particular view of the bootstrap. Secondly, it gives an account of Edgeworth expansion. It is aimed at a graduate-level who have some exposure to the methods of theoretical statistics.
I'm still working my way through this one a fourth time. Textbook treatments of the Edgeworth expansion are hard to find, and the connection to the more familiar Bootstrap helps the reader to understand both topics. The examples good be better, or at least more realistic.