In corporate finance, the early prediction of financial distress is considered more important as another occurrence of business risks. The study presents a review of literature for early prediction of financial bankruptcy. It contributes to the formation of a systematic review of the literature regarding previous studies done in the field of bankruptcy. It addresses two most commonly used financial distress prediction models, i.e. multivariate discriminant analysis and logit. Models are discussed with their advantages and disadvantages. After methodological review, it seems that logit regression model (LRM) is more advantageous than multivariate discriminant analysis (MDA) for better prediction of financial bankruptcy. However, accurate prediction of bankruptcy is beneficial to improve the regulation of companies, to form policies for companies and to take any precautionary measures if any crisis is about to come in future.

Financial distress is a debatable issue among the researchers especially in developing economies. This study investigates the significant financial indicators for five manufacturing sectors listed on Pakistan Stock Exchange (PSX). Sample consists of 35 bankrupt and 156 non-bankrupt companies from textile, cement, sugar, technology and communication and power generation and distribution sectors. Study uses two data sets from 2005 to 2013 for estimation sample and 2014 to 2016 for holdout sample. Logistic regression analysis comprises of sixteen financial ratios under respective indicators i.e. profitability, liquidity, leverage, asset efficiency and size. Findings of the study reveal six significant financial ratios for each indicator i.e. return on equity (ROE), quick ratio, current ratio, shareholder's equity to total assets, sales to current assets and natural log of total assets. Results show overall model accuracy of 89.6 percent for estimation sample while 92.2 percent for holdout sample which indicates model consistency with better financial distress prediction power in the context of Pakistan.

No abstract