This paper aims to develop a corporate financial distress model for Indian listed companies using financial and non-financial parameters by using a conditional logit regression technique.,This study used a sample of 96 companies, of which 48 were declared sick between 2014 and 2016. The sample was divided into a training sample and a testing sample. The variables for the study included nine financial variables and four non-financial variables. The models were developed using financial variables alone as well as combining financial and non-financial variables. The performance of the test sample was measured with confusion matrix, sensitivity, specificity, precision, F-measure, Types 1 and 2 error.,The results show that models with financial variables had a prediction accuracy of 85.19 and 86.11 per cent, whereas models with a combination of financial and non-financial variables predict with comparatively better accuracy of 89.81 and 91.67 per cent. Net asset value, long-term debt–equity ratio, return on investment, retention ratio, age, promoters holdings pledged and institutional holdings are the critical financial and non-financial predictors of financial distress.,This study contributes to the financial distress prediction literature in different ways. First, there have been, until now, few studies in the area of financial distress prediction in the Indian context. Second, business failure studies in the past have used only financial variables. The authors have combined financial and non-financial variables in their model to increase predictive ability. Thirdly, in most earlier studies, variable institutional holdings were found to affect financial distress negatively. In contrast, the authors found this parameter to be positively significant to the financial distress of the company. Finally, there have hitherto been few studies that have used promoter holdings pledged (PHP) or pledge ratio. The authors found this variable to influence business failure positively.

中文翻译：

---

使用财务和非财务变量对公司财务困境进行建模

本文旨在通过条件logit回归技术为印度上市公司使用财务和非财务参数建立公司财务困境模型。该研究使用了96家公司的样本，其中48家在2014年至2016年期间被宣布为有病。样本分为训练样本和测试样本。该研究的变量包括九个财务变量和四个非财务变量。仅使用财务变量以及合并财务和非财务变量来开发模型。用混淆矩阵，敏感性，特异性，精密度，F度量，1型和2型误差对测试样品的性能进行了测量。结果表明，具有财务变量的模型的预测准确度为85.19％和86.11％，而结合了金融变量和非金融变量的模型预测的相对精度更高，为89.81％和91.67％。资产净值，长期债务权益比率，投资回报率，保留率，年龄，发起人的抵押品和机构资产是财务困境的关键财务和非财务预测指标。该研究为财务困境预测文献做出了贡献以不同的方式。首先，到目前为止，在印度背景下，关于财务困境预测的研究很少。第二，过去的业务失败研究仅使用财务变量。作者在其模型中结合了财务和非财务变量以提高预测能力。第三，在较早的研究中，发现可变的机构持股会对财务困境产生负面影响。相反，作者发现此参数对公司的财务困境具有积极意义。最后，迄今为止，很少有研究使用启动子持股（PHP）或质押比率。作者发现此变量可对业务失败产生积极影响。