Dissertation > Excellent graduate degree dissertation topics show

Evaluation of credit risk portfolio of enterprise groups

Author: LiuWenRui
Tutor: ZhouZongFang
School: University of Electronic Science and Technology
Course: Finance
Keywords: Enterprise Group credit risk Hybrid Model BP neural network Logistic Regression
CLC: F272
Type: Master's thesis
Year: 2011
Downloads: 68
Quote: 3
Read: Download Dissertation

Abstract


Enterprise Group has the characteristic of large-scare, and operating cross-industry and cross-zone. Also, there is the complicated affiliated transaction existed in it. For the unusual speed of development, now Enterprise Group is having the pivotal influence on the global economic. The bankrupt of famous Enterprise Group like Lehman Brothers and General Motor, took amount of money involved, not only has affected banks, but also impacted the whole Financial industry even global economic deeply. And because of this tremendous influence on economic field and even social field, the credit stand of Enterprise Group is becoming the focus. Then how to prevent Credit Risk of Enterprise Group effectively and how to evaluate the credit stand reasonably is becoming the tendency and future of research, which is also having practical significance in social.According to the characteristics of Enterprise Group, this paper chooses affiliated transaction as characteristic index of the credit risk of Enterprise Group. Then chooses correlative Chinese listed Enterprise Group over the 2004-2008 periods as study sample, applies Logistic Regression and BP neural network to build Single Model and Hybrid Model. The principal results are as follows:1. Researches the characteristics, current situation and reasons of Credit Risk in Enterprise Group. According to the characteristics of Enterprise Group, this paper leads into affiliated transaction index and other indexes as the original assessment indexes.2. Reduces original indexed based on Rough Set Theory and Genetic Algorithms, and gets 11 indexes left. Then divides sample into separate levels for defining the training sample and testing sample, and conducts data into unitary data.3. Builds the credit risk assessment models of Enterprise Group separately by applying Logistic Regression and BP neural network, according to the training sample. Then we testes and gets accuracy of those Single Models.4. Builds Hybrid Model which is using Logistic Regression to establish risk assessment model. Then we put these generative scores and other data into BP neural network which are as Explanatory variables. We Testes and gets accuracy of this Model.5. Compares the analysis of these three models, the results shows that, when compares the two Single Model, BP neural network is more accurate on credit risk assessment of enterprise group. But when compares the Single Model and Hybrid Model, the second model is more accurate. Especially, the superiority is much obvious in the first fallacious rate, which reflects the real credit loss. Thus, the research results show that the accuracy of Hybrid Model is better than the Single Model in credit risk assessment of Enterprise Group.

Related Dissertations

  1. Multiple ANN/HMM Hybrid Used in Speech Recognition,TN912.34
  2. Research on Feature Extraction and Classification of Tongue Shape and Tooth-Marked Tongue in TCM Tongue Diagnosis,TP391.41
  3. Research on Visual Servo System of Mechanical ARM,TP242.6
  4. Municipal tourism land use planning environmental impact assessment,X820.3
  5. Study on Taste Characteristic of Taste Peptide Enzymatic Production from Oyster Base on A Neural Network Method,TS254.4
  6. The Research on Evaluation of Living Status Systems of Expressway Relocated People,D523
  7. The Research on a Predictive Model of Depression among College Students,B842.6
  8. MIMIC, MH and LR methods in the analysis of the project functional differences comparative study,B841.7
  9. Vocational human resource management incentive mechanism,G717.2
  10. Colorectal surgery surgical site infection Logistic regression analysis,R656.7
  11. Chronic pharyngitis secondary school teachers in Kunming Quantitative Evaluation of Risk Factors,R766.14
  12. Mine Risk Information Integration and Intelligent Early Warning,X936
  13. Publishing credit Credit Risk Control Problem,G231-F
  14. Research of Virus Detection Methods Based on Multiple Anti-virus Softwares Collaboration,TP309.5
  15. Research of Orange Quality Classification Technology Based on Computer Vision,TP391.41
  16. The Research of the Internal Control Issues of Enterprise Group Based on Corporate Governance,F275
  17. Optimization Study on Gating System and Molding Process Parameters of Injection Mold Based on Simulation,TQ320.662
  18. A Study on Early Warning of Grassland Fire Disaster Risk in Hulunbeier of Inner Mongolia,S812.6
  19. Research on Credit Risk of Suzhou Listed Companies Base on Logistic Regression Analysis,F224;F832.4
  20. Application Research of Genetic Neural Network in Surface Water Evaluation,X824
  21. The Development of China’s Rural Micro-credit Risk Control Study,F832.4

CLC: > Economic > Economic planning and management > Enterprise economy > Enterprise planning and management decision - making
© 2012 www.DissertationTopic.Net  Mobile