Key features include: * A unified approach to statistical estimation emphasising the analogy (or bootstrap) principle * An introduction to bootstrap and jackknife methods for assessing the accuracy of an estimator * Detailed discussion of nonparametric methods for estimating density and regression of functions * Emphasis on diagnostic procedures and on prediction criteria for evaluating the results fo statistical analysis * An introduction to linear exponential family and generalized linear models * A thorough discussion of robustness in statistical sense

Preface. Notation. Regression Models. Sampling. Time Series. Point Estimation. Statistical Accuracy and Hypothesis Testing. The Classical Linear Model: Estimation. Violations of the Ideal Conditions for OLS. Diagnostics Based on the OLS Estimates. The Classical Linear Model: Hypothesis Testing. Asymptotic Properties of Least Squares Methods. The Instrumental Variables Method. Linear Models for Panel Data. Linear Simultaneous Equation Models. Nonparametric Methods. M-Estimators. Adaptive and Robust Regression Estimators. Models for Discrete Responses. Models for Truncated and Censored Data. References. Appendix A: Review of Linear Algebra. Appendix B: Methods of Numerical Maximization. Appendix C: Review of Probability. Appendix D: Elements of Asymptotic Theory. Index.

Franco Peracchi is a Professor of Econometrics at Tor Vergata University in Rome. he received an MSc in Econometrics from the London School of Economics in 1983 and a PhD in Economics from Princeton University in 1987. His research interests include econometric theory and methods, nonparametric and robust statistical methods, and labour economics. His work has been published in leading econometric and statistical journals.