|
|
55 of 59 people found the following review helpful:
3.0 out of 5 stars
Why won't anyone just call it the Battle of the Sexes?, August 24, 2000
I used this book for the second half of a principles of micro. course, to supplement the shoddy, one-chapter-on-the-Prisoners'-Dilemma treatment found in most principles textbooks. Everything said here should be interepreted according to this. From this instructor, the book gets points for being the best one available for teaching low-level undergraduates, but it could have been a LOT better. For example:The notation and terminology are in many cases non-standard, and tend to change from chapter to chapter. The BoS is the Battle of Cultures (though this is not the first book to mess with this game). Chicken is a Game of Assurances, except in Ch. 10. SPE are (quasi-)formally described in Ch. 6, but they are actually introduced in Ch.4, where they are called Rollback Equilibria. Many times, I would have to tell students, "this is what your book calls a..." The authors use confusing and convoluted examples to motivate concepts. For example, it takes a confusing, two-page story about advertizing in a political race to motivate study of sequential-move games. A simple entry-deterrence story gets the point across. Also on this point, sequential-move games appear before simultaneous-move ones. I reversed this, in part ot be able to show that the set of SPE is merely a subset of the set of NE (again, using the entry-deterrence story). In fact, there's no real attempt to relate many of (seemingly unrelated) concepts to one another, as equilibrium refinements, each of which conforms to some intuitive concept of the "right" way of playing a given game. The disucssion of the special case of two-person, zero-sum games, introducing pre-Nash notation and solution concepts is merely confusing for the uninitiated. I see no reason that anyone not yet in graduate school should have to know the min-max theorem. In some ways, the books seems to suffer from over- and under-reach at the same time. The subject of infinitely repeated games gets two pages on TFT and Grim strategies in a repeated Prisoners' Dilema. There's no real discussion of rationalizability, or Bayesian games; many important concepts are smooshed into a couple of chapters, like they're being swept under the rug. There IS, however, a chapter on evolutionary games, and a (math-free) chapter on auctions. Again, these are points that, I think, led to undue confusion, and required undue effort to counteract. However, I don't mean to be unduly harsh. I'm not suggesting that the authors should merely have mimeographed Fudenberg & Tirole, and whited-out the math. This is a useful book, ahead (as far as I know) of other treatments appropriate for students at this level. But it could have been much better.
|
by Kenneth A. Shepsle
by Martin J. Osborne
