A Generic Model Of Predicting Probability Of Success-Distress Of An Organization: A Logistic Regression Analysis

Many bankruptcy prediction models have been created over the years using a mix of variables derived mostly from accrual-based accounting statements and were industry specific. The primary issue with using a model comprised of accrual-based variables is that firm management can manipulate different components and make the balance sheet and income statement misleading (Wanuga 2006). Thus, firms appear financially healthy yet unable to meet the day-to-day cash flow needs of the firm; these financial issues are less likely to be hidden in the cash flow statement (Sharma 2001). In this study, we use a binary regression model with theoretically supported variables obtained from the cash flow statement to forecast firm success versus distress. Of particular interest, we examine firms representing 85 industries using firm data during and immediately following the greatest recession in United States history (Fieldhouse 2014; Lee 2014). The model is generic in the sense that it can be used to predict the probability of success-distress of any entity using the three major financial statements. We find that the overall model correctly classifies organizations 90.290 percent of the time.