This study compares samples of publicly and privately held bank holding companies to examine whether the high frequency of small earnings increases relative to small earnings decreases reported by public firms is attributable to earnings management. We expect public banks' shareholders to be more likely than private banks' shareholders to rely on simple earnings-based heuristics in evaluating firm performance, so we expect public banks to have more incentives to report steadily increasing earnings. Consistent with this expectation, we find that relative to private banks, public banks: (1) report fewer small earnings declines, (2) are more likely to use the loan loss provision and security gain realizations to eliminate small earnings decreases, and (3) report longer strings of consecutive earnings increases. These results suggest that the asymmetric pattern of more small earnings increases than decreases, first documented by Burgstahler and Dichev (1997), is attributable to earnings management and is not simply a reflection of the underlying distribution of earnings changes.

No abstract

We develop a state-of-the-art fraud prediction model using a machine learning approach. We demonstrate the value of combining domain knowledge and machine learning methods in model building. We select our model input based on existing accounting theories, but we differ from prior accounting research by using raw accounting numbers rather than financial ratios.We employ one of the most powerful machine learning methods, ensemble learning, rather than the commonly used method of logistic regression. To assess the performance of fraud prediction models, we introduce a new performance evaluation metric commonly used in ranking problems that is more appropriate for the fraud prediction task. Starting with an identical set of theory-motivated raw accounting numbers, we show that our new fraud prediction model outperforms two benchmark models by a large margin: the Dechow et al. logistic regression model based on financial ratios, and the Cecchini et al. support-vector-machine model with a financial kernel that maps raw accounting numbers into a broader set of ratios. JEL codes: C53; M41

We examine whether transient institutional investors (i.e., institutions that trade actively to maximize short-term profits) have information that allows them to predict a break in a string of consecutive quarterly earnings increases and thereby avoid the economically significant negative stock price response associated with the break announcement. We show that transient institutions predict the break at least one quarter in advance of the break quarter. We also provide evidence that is consistent with transient institutions obtaining information regarding the impending break from private communications with management. Copyright University of Chicago on behalf of the Institute of Professional Accounting, 2004.

No abstract