A Course in Simulation (Sheldon M. Ross)

tationally intensive, and relevant algorithms are given. An enormous body of material is touched on or alluded to in the 10 chapters and two appendices: the ACE algorithm, projection pursuit, regression trees, monotonic and unimodal smoothing, L-smoothing, Rsmoothing, M-smoothing, various forms of cross-validation, bootstrap techniques, confidence and variability bands, recursive smoothing, k-nearest neighbor estimates, and much more. Now for the down side. Occasionally, the book reads like a dictionaryma string of ordered, globally undigested morsels. The reader is left to draw his or her own conclusions about what technique to use in a given situation. The hurried reader looking for handy rules of thumb, stemming from thoughtful comparisons of the various alternatives, will be disappointed. This maywell be inevitable given the present state of the subject. To follow such a policy is certainly the safest for the author. Certainly, this book is the place to look for ideas and to obtain a feel for the directions of past and present research.