Bankruptcy Prediction Model for U S Telecommunication Network Providers: Study of Financial Data

  • 馬 山杰

Student thesis: Master's Thesis


In this study binary logistic regression and binary quantile regression are used to come up with bankruptcy prediction model for telecommunication carriers’ bankruptcy illiquidity The purpose of this research is to find out key determinants of regarding insolvency in telecommunication industry and to evaluate 2 different regressions’ performance Operating margin Receivables turnover Average collection period and Total asset are included in proposed model Research results show that proposed model with 4 explanatory variables are highly useful to classify and predict firms as bankrupted and survived ROC and CAP curve indicates that model by binary logistic regression correctly discriminate 94% and 84% respectively and slightly better than binary quantile regression But binary quantile regression demonstrates more complete estimates for different quantile levels and shows how explanatory variables’ estimations’ move in relative to probability of bankruptcy
Date of Award2015 Feb 4
Original languageEnglish
SupervisorAnn Shawing Yang (Supervisor)

Cite this