The second edition of this acclaimed graduate text provides a unified treatment of the analysis of two kinds of data structures used in contemporary econometric research: cross section data and panel data. The book covers both linear and nonlinear models, including models with dynamics and/or individual heterogeneity. In addition to general estimation frameworks (particularly methods of moments and maximum likelihood), specific linear and nonlinear methods are covered in detail, including probit and logit models, multinomial and ordered choice models, Tobit models and two-part extensions, models for count data, various censored and missing data schemes, causal (or treatment) effect estimation, and duration analysis. Control function and correlated random effects approaches are expanded to allow estimation of complicated models in the presence of endogeneity and heterogeneity.
