Search results

While there is a substantial literature in labor economics and microeconometrics directed toward endogenous causal effects, causal effects have received relatively limited attention in accounting. This volume builds on econometric foundations, including linear, discrete choice, and nonparametric regression models, to address challenging accounting issues characterized by microeconomic fundamentals and equilibrium reporting choices. Both classical and Bayesian strategies for identifying and estimating accounting treatment effects are discussed extensively. This distinctive resource for researchers and students explores interactions among theory, data, and model specification considerations, and complements contemporary econometrics and statistics, as well as accounting. TOC:Preface. 1. Introduction.- 2 . Accounting choice.- 3. Linear models.- 4. Loss functions and estimation.- 5. Discrete choice models.- 6. Nonparametric regression.- 7. Repeated-sampling inference.- 8. Overview of endogeneity.- 9. Treatment effects: ignorability.- 10. Treatment effects: IV.- 11. Marginal treatment effects.- 12. Bayesian treatment effects.- 13. Informed priors.- Appendix: Asymptotic theory.- Bibliography.- Index.

One way to develop an appreciation for the power and beauty of doubleentry is to derive consistent transaction vectors, i.e., start with financial statements and derive the transaction amounts that could have generated the statements. Working backward to uncover transactions (the inverting exercise) complements the more traditional approach of working forward from transactions to financial statements. In addition, the approach highlights a fundamental accounting activity: aggregation. Financial statements summarize a firm's transactions using only a relatively small number of account balances. A consequence of aggregation is that there are infinite consistent transaction vectors. However, because these infinite solutions are all prepared in accordance with double-entry, they are linked to each other in a systematic fashion. We show how a directed graph representation of the accounting system can be used as a parsimonious means of characterizing all consistent transaction vectors. As the number of accounts and transactions are increased, the inverting exercise becomes tedious if one does not make use of the directed graph. To emphasize this point, the inverting exercise is conducted using a set of published financial statements. We also discuss the issue of picking the most likely transaction vector from the set of consistent transaction vectors. The authors have used this note in undergraduate, M.B.A., and Ph.D. classes. For the Ph.D. class, the note has been supplemented by the more rigorous analyses in the linear algebra literature and the accounting literature.