Lennart Ljung

Lennart Ljung is one of the leading researchers in this area. However, the book has many typos. I noticed three in the definition of convergence (page 11) alone! So it reads like the minutes of a conference typed up on the fly by a secretary who knew nothing about stochastic approximation and had been given the overwhelming task of typing in real time. The problem with this (as with Azariadis's textbook) is that a methods book that one knows has lots of errors is of very diminished usefulness, since one doesn't know what to trust. Aargh!… (mais)