Applicability of logistic regression analysis in predicting financial distress for firms listed at the Nairobi securities exchange

Abstract

Logistic regression combines independent variables to estimate the probability that a particular event will occur, i.e. a subject will be a member of one of the groups defined by the dichotomous dependent variable. If the probability for group membership in the modeled category is above some cutoff point, the subject is predicted to be a member of the modeled group. ]f the probability is below the cutoff point, the subject is predicted to be a member of the other group. This study assesses the probability of firm failure, a year before failure, using logistic regression model that was developed by Ohlson (1980). The study utilized secondary data collected from Capital Markets Authority and Nairobi Securities Exchange. The required data was collected from financial statements of a sample of sixteen companies; ten of which were in good financial health and six of which were financially distressed. The study covered a range of 14 years from 1997 to 2011. The results of this study show that logistic regression analysis is applicable in 9 out of 10 firms analyzed which indicates a 90% successful application of the Ohlson (1980) model used in the study. The model is found to be successful in predicting business failure one year before it occurs. The study concludes that the logistic regression analysis model developed by Ohlson (1980) is applicable in predicting financial failure of companies and is a useful tool for investors in the Kenyan market. This study contributes to the literature by expanding the application of Ohlson (1980) logit model to Kenya publicly listed companies. It provides applicable measures for predicting firm delisting events in Kenya's stock markets. The study recommends that the assessment be extended is to test bankruptcy prediction models to the non-listed, relatively smaller turnover sized firms where the incidence of business failure is greater than larger corporations.