Description

How to learn both applied statistics (econometrics) and free, open-source software R? This book allows students to have a sense of accomplishment by copying and pasting many hands-on templates provided here.

The textbook is essential for anyone wishing to have a practical understanding of an extensive range of topics in Econometrics. No other text provides software snippets to learn so many new statistical tools with hands-on examples. The explicit knowledge of inputs and outputs of each new method allows the student to know which algorithm is worth studying. The book offers sufficient theoretical and algorithmic details about a vast range of statistical techniques.

The second edition's preface lists the following topics generally absent in other textbooks. (i) Iteratively reweighted least squares, (ii) Pillar charts to represent 3D data. (iii) Stochastic frontier analysis (SFA) (iv) model selection with Mallows' Cp criterion. (v) Hodrick-Prescott (HP) filter. (vi) Automatic ARIMA models. (vi) Nonlinear Granger-causality using kernel regressions and bootstrap confidence intervals. (vii) new Keynesian Phillips curve (NKPC). (viii) Market-neutral pairs trading using two cointegrated stocks. (ix) Artificial neural network (ANN) for product-specific forecasting. (x) Vector AR and VARMA models. (xi) New tools for diagnosing the endogeneity problem. (xii) The elegant set-up of k-class estimators and identification. (xiii) Probit-logit models and Heckman selection bias correction. (xiv) Receiver operating characteristic (ROC) curves and areas under them. (xv) Confusion matrix. (xvi) Quantile regression (xvii) Elastic net estimator. (xviii) generalized Correlations (xix) maximum entropy bootstrap for time series. (xx) Convergence concepts quantified. (xxi) Generalized partial correlation coefficients (xxii) Panel data and duration (survival) models.

Contents:

• Production Function and Regression Methods Using R
• Univariate Time Series Analysis with R
• Bivariate Time Series Analysis Including Stochastic Diffusion and Cointegration
• Utility Theory and Empirical Implications
• Vector Models for Multivariate Problems
• Simultaneous Equation Models
• Limited Dependent Variable (GLM) Models
• Consumption and Demand: Kernel Regressions and Machine Learning
• Single, Double, and Maximum Entropy Bootstrap and Inference
• Generalized Least Squares, VARMA, and Estimating Functions
• Box–Cox, Loess, Projection Pursuit, Quantile and Threshold Regression
• Miscellany: Dependence, Correlations, Information Entropy, Causality, Panel Data, and Exact Stochastic Dominance

Readership: Undergraduate and graduate students of economics and econometrics, applied statisticians and finance professionals.
Key Features:

• Uniquely comprehensive coverage, including many very recently developed topics
• Includes software snippets (that help learn the R language) for readers who are not interested in economics examples