Hayashi's Econometrics promises to be the next great synthesis of modern econometrics. It introduces first year Ph.D. students to standard graduate econometrics material from a modern perspective. It covers all the standard material necessary for understanding the principal techniques of econometrics from ordinary least squares through cointegration. The book is also distinctive in developing both time-series and cross-section analysis fully, giving the reader a unified framework for understanding and integrating results.
I have some misgivings about the field of econometrics, fundamental elements are missing, and traditionally even if GMM estimation, is within the realm of admissable methods in statistics, it is quite fringe. This is also true of MA, ARMA and ARIMA for me, because in the case of economics they are not backed by fundamentals, nor do they elucidate them, given a theoretical mismatched. This is on some level, the same misgivings I have for Robert Lucas Jr.'s attempt to bridge micro to macro, through the use of statistical methods. There is some intuition or concept gap, where certain things are not being bridged. At least from what limited I have read. Perhaps, this is worthy of some further exploration.
A great introduction to graduate-level econometrics, especially because it ties almost everything under a generalized method of moments (GMM) framework (OLS, GLS, IV, 2SLS, SUR, 3SLS, random and fixed effects models, etc.), something which I failed to appreciate until the end of the first year of my graduate econometrics curriculum.
The one criticism I have is that occasionally the notation was hard to follow, since most textbooks reserve X for (possibly endogenous) regressors and Z for instruments, whereas Hayashi prefers the complete opposite.