Reddit reviews Estimation and Inference in Econometrics

This is indeed a good book, but it really has an algebraic approach to the field. It is def. useful, but lacks a lot in treatment of probabilistic aspects. There is a huge focus on Gauss-Newton regressions which you might not need now. Bootstrapping is also very prominently displayed (research area of the authors).

If you like the style though, consider http://www.amazon.com/Estimation-Inference-Econometrics-Russell-Davidson/dp/0195060113 by the same authors. It gives a better treatment.