Empirical Research in Accounting: Tools and Methods (Chapman and Hall/CRC Series on Statistics in Business and Economics)

This textbook provides the foundation for a course that takes PhD students in empirical accounting research from the very basics of statistics, data analysis, and causal inference up to the point at which they conduct their own research. Starting with foundations in statistics, econometrics, causal inference, and institutional knowledge of accounting and finance, the book moves on to an in-depth coverage of the core papers in capital market research. The latter half of the book examines contemporary approaches to research design and empirical analysis, including natural experiments, instrumental variables, fixed effects, difference-in-differences, regression discontinuity design, propensity-score matching, and machine learning. Readers of the book will develop deep data analysis skills using modern tools. Extensive replication and simulation analysis is included throughout. Key Features: Extensive coverage of empirical accounting research over more than 50 years. Integrated coverage of statistics and econometrics, institutional knowledge, and research design. Numerous replications and a dozen simulation analyses to immerse readers in papers and empirical analysis. All tables and figures in the book can be reproduced by readers using included code. Easy-to-use templates facilitate hands-on exercises and introduce reproduceable research concepts. (Solutions available to instructors.) Cover Half Title Series Page Title Page Copyright Page Contents Preface I. Foundations 1. Introduction 1.1. Structure of the book 1.2. Setting up your computer 2. Describing data 2.1. Introduction to R 2.2. Exploring data 2.3. Basic data analysis and statistics 2.4. Reproducible research 2.5. Further reading 2.6. Exercises 3. Regression fundamentals 3.1. Introduction 3.2. Running regressions in R 3.3. Frisch-Waugh-Lovell theorem 3.4. Further reading 4. Causal inference 4.1. Econometrics 4.2. Basic causal relations 4.3. Causal diagrams: Formalities 4.4. Discrimination and bias 4.5. Causal diagrams: Application in accounting 4.6. Further reading 5. Statistical inference 5.1. Some observations 5.2. Data-generating processes 5.3. Hypothesis testing 5.4. Differences in means 5.5. Inference with regression 5.6. Dependence 5.7. Further reading 6. Financial statements: A first look 6.1. Setting up WRDS 6.2. Financial statement data 6.3. Exercises 7. Linking databases 7.1. Firm identifiers 7.2. The CRSP database 7.3. Linking CRSP and Compustat 7.4. All about CUSIPs 8. Financial statements: A second look 8.1. Core attributes of financial statements 8.2. Balance sheets 8.3. Within-statement articulation 8.4. Across-statement articulation 8.5. Missing R&D 9. Importing data 9.1. Reading (seemingly) non-tabular data 9.2. Extracting data from messy formats 9.3. Further reading II. Capital Markets Research 10. FFJR 10.1. Efficient capital markets 10.2. Stock splits 10.3. Dividend policy 10.4. Replication of FFJR 10.5. Capital markets research in accounting 11. Ball and Brown (1968) 11.1. Principal results of Ball and Brown (1968) 11.2. Replicating Ball and Brown (1968) 12. Beaver (1968) 12.1. Market reactions to earnings announcements 12.2. A re-evaluation of Beaver (1968) 12.3. Discussion questions 13. Event studies 13.1. Overview 13.2. The modern event study 13.3. Event studies and regulation 14. Post-earnings announcement drift 14.1. Fiscal years 14.2. Quarterly data 14.3. Time-series properties of earnings 14.4. Post-earnings announcement drift 15. Accruals 15.1. Sloan (1996) 15.2. Measuring accruals 15.3. Simulation analysis 15.4. Replicating Sloan (1996) 15.5. Accrual anomaly 16. Earnings management 16.1. Measuring earnings management 16.2. Evaluating measures of earnings management 16.3. Power of tests of earnings management III. Causal Inference 17. Natural experiments 17.1. Randomized experiments 17.2. Natural experiments 17.3. Recognition versus disclosure 17.4. Michels (2017) 17.5. Discussion questions 18. Causal mechanisms 18.1. John Snow and cholera 18.2. Smoking and heart disease 18.3. Causal mechanisms in accounting research 18.4. Discussion questions 19. Natural experiments revisited 19.1. A replication crisis? 19.2. The Reg SHO experiment 19.3. Analysing natural experiments 19.4. Evaluating natural experiments 19.5. The parallel trends assumption 19.6. Indirect effects of Reg SHO 19.7. Statistical inference 19.8. Causal diagrams 19.9. Causal mechanisms 19.10. Two-step regressions 20. Instrumental variables 20.1. The canonical causal diagram 20.2. Estimation 20.3. Reasoning about instruments 20.4 “Bullet-proof” instruments 20.5. Causal diagrams: An application 20.6. Further reading 20.7. Discussion questions and exercises 21. Panel data 21.1. Analysis of simulated data 21.2. Voluntary disclosure 21.3. Further reading 22. Regression discontinuity designs 22.1. Sharp RDD 22.2. Fuzzy RDD 22.3. Other issues 22.4. Sarbanes-Oxley Act 22.5. RDD in accounting research 22.6. Further reading 22.7. Discussion questions IV. Additional Topics 23. Beyond OLS 23.1. Complexity and voluntary disclosure 23.2. Generalized linear models 23.3. Application: Complexity and voluntary disclosure 23.4. Further reading 23.5. Appendix: Maintaining a repository of SEC index files 24. Extreme values and sensitivity analysis 24.1. Leone et al. (2019) 24.2. Call et al. (2018) 24.3. Sensitivity analysis 24.4. Appendix: Simulation study from Leone et al. (2019) 25. Matching 25.1. Background on auditor choice 25.2. Simulation analysis 25.3. Replication of Lawrence et al. (2011) 25.4. DeFond et al. (2017) 25.5. Further reading 25.6. Discussion questions 26. Prediction 26.1. Prediction in research 26.2. Predicting accounting fraud 26.3. Model foundations 26.4. Performance metrics 26.5. Penalized models 26.6. Ensemble methods 26.7. Testing models 26.8. Further reading 26.9. Discussion questions and exercises V. Appendices A. Linear algebra A.1. Vectors A.2. Matrices A.3. The OLS estimator A.4. Further reading B. SQL primer B.1. What are SQL and dplyr? B.2. SQL terms SELECT and FROM B.3. SQL WHERE B.4. SQL ORDER BY B.5. SQL approach to mutate() B.6. SQL GROUP BY and aggregates C. Research computing overview C.1. Languages C.2. Data management D. Running PostgreSQL D.1. Setting up a personal server D.2. Setting up a shared server E. Making a parquet repository E.1. Data management approaches E.2. Organizing data E.3. Canonical WRDS data E.4. Converting WRDS data to parquet E.5. Working with parquet files E.6. Creating a parquet library References Index