This text provides an introduction to this complex topic. It does this by providing the student with an analytical and an intuitive understanding of the classical linear regression model. Mathematical notation is kept simple and step-by-step verbal explanations of mathematical proofs are provided to facilitate a full understanding of the subject. The text also contains a large number of practical exercises for students to follow up and practice what they have learnt. This new edition has been substantially updated and revised with the inclusion of new material on specification tests, binary choice models, tobit analysis, sample selection bias, nonstationary time series, and unit root tests and basic cointegration. The new edition is also accompanied by a website with Powerpoint slideshows giving a parallel graphical treatment of topics treated in the book, cross-section and time series data sets, manuals for practical exercises, and lecture note extending the text.