Use this URL to cite or link to this record in EThOS:
Title: Business failure prediction model for the construction industry using financial ratios and entropy measures with discriminant analysis
Author: Wu, Hsu-Che
ISNI:       0000 0001 3573 300X
Awarding Body: University of Warwick
Current Institution: University of Warwick
Date of Award: 2007
Availability of Full Text:
Full text unavailable from EThOS. Restricted access.
Please contact the current institution’s library for further details.
The purpose of this research is to examine empirically the effectiveness of entropy measures derived from information theory combined with discnminant analysis in the prediction ofconstruction business failure. The research will use data derived from the construction industry as business failure is an extremely disruptive force, and a key factor in the prequalification appraisal of contractors in the construction industry (Kangari, Farid et al. 1992; Russell 1992; Yang 1997). The study has used Taiwan construction industry data for empirical analysis. In this research, the model is based on a financial distress definition of business failure. Previous research has shown that the decomposition measure for financial statements has the power to discriminate with respect to failed and non-failed firms. However, the predictive ability is weaker than with just financial ratios (Hamer 1980). This work I explores the transformation of financial ratios to information decomposition measures using information theory. The transformation can' adjust the data to be naturally dynamic. The accumulative dynamic information measures offinancial ratios ar{, then compared with a static financial ratios model in terms oftheir failure prediction ability. It appears that researchers have never utilized information measures of financial ratio to analyse whether this can improve predictive ability. Thus, this new work modifies discriminant analysis with the information measures offinancial ratios. Therefore, this study attempts to bridge this gap in earlier studies. The method devised contributes to measuring the risk (financial distress) of companies in the construction industry or contractors on tender lists. Financial data was gathered for 15 failures and 30 non-failure examples from Taiwanese construction companies between 1997 and 2002.The research uses statistical methodology to test the most important hypotheses 'The predictive ability using dynamic information measures and discriminant analysis is not more accurate than usingjinancial ratios and discriminant analysis'. In conclusion, the author provides evidence to reject the hypotheses. The results reflect a remarkable effect ofthe model using dynamic infonnation measures on perfonnance in the three different models (prediction Model with Financial Ratios, Prediction Model with Static Infonnation Measures, and Prediction Model with Dynamic Infonnation Measures). In the research, the SAS (Business Intelligence software) is used to develop the discriminant models for the study. In future research, the author suggests the combining of the agent technology concept and the tool to extend a real time dynamic prediction model. An agent as an early warning system may be the key to the constant monitoring ofthe financial condition of a company.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available