Use this URL to cite or link to this record in EThOS:
Title: Risk management in SMEs using data mining methods with financial and non-financial indicators
Author: He, Chenyang
ISNI:       0000 0004 7961 7520
Awarding Body: Brunel University London
Current Institution: Brunel University
Date of Award: 2018
Availability of Full Text:
Access from EThOS:
Access from Institution:
As the economy develops, the importance of risk management increases significantly. Newly-developed data mining techniques have also provided an extended range of information for decision-makers and scholars about the Risk Management (RM) process. This study thus assessed the use of financial and non-financial indicators in the RM process based on a Business Intelligence (BI) approach and using data mining (DM) methodology. Its assessment focused on the selection of Key Risk Indicators (KRIs) among the various risk indicators for performance measurement and risk control. This study used a sample of 853 Chinese SMEs listed on the Shenzhen Stock Exchange. After comparison of LR, GA, NN, and CHAID, CHAID was found to be the most suitable mode, as it incorporates both financial and non-financial indicators and is also able to provide roadmaps to improve RM performance. This study also used a BI approach to quantify and standardise information from government reports and firms' annual reports to better generalise the available information for non-financial indicators. Four different types of risks were considered, following the enterprise risk management (ERM) framework, and using CHAID as the underlying method, the threshold values and roadmaps of the KRIs were thus identified. This study thus provides an integrated method for the risk management process in SMEs by using both financial and non-financial information generalised using a BI approach with the DM process. The critical contribution of this study is its combination of the DM process and RM process, which also allowed examination of the usefulness of non-financial indicators in the RM process with the ERM framework. Additionally, it provides practical guidance for using a BI approach for capturing information and transferring data.
Supervisor: Lu, K. ; Althonayan, A. Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available
Keywords: Key risk indicators ; Machine learning ; Enterprise risk management ; Early warning system ; Business intelligence