In the wake of accounting frauds and scams in the corporate sector, predicting abnormal or fraudulent behavior is as important as investigating a fraud. In this respect, this theme of “Early warning system” designs an early warning measure of financial distress based on the qualitative information present in the corporate annual reports (Directors report, Audit report and Notes to Accounts). A systemic financial distress prediction process based on the tone of annual report text information is constructed and quantifies both positive (sunshine index) and negative sentiments (fear index) in the annual report’s language without using any accounting information. The study used 4000 annual reports obtained from MCA which had sufficient textual information to compute distress index for each firm. Based on the distress index 50 % of the firms were identified as distressed firms.
Keywords
Bag of words approach, feature extraction, feature engineering, fear index, sunshine index, distress intensity, Altman EMS
This theme of "Early Warning System" throws light on the issue of earnings management prevalent in the Indian corporate sector which has received considerable attention from academics and regulators recently. The study develops an Earnings management Score (EMS), a metric to define the magnitude of opportunistic earnings management, for Indian non-financial firms using their financial statement- balance sheet, income statement and cash flow statement. EMS for 5782 firms belonging to 37 industries is generated. Also, a comparative study of earnings management across different sector using Bedford’s test is done along with identification of key variables susceptible to earnings manipulation.