An MDA-Based Credit Scoring Model for Malaysian SMEs: Evaluating the Incremental Contribution of Non-Financial Indicators

Lim, Sze Kei (2026) An MDA-Based Credit Scoring Model for Malaysian SMEs: Evaluating the Incremental Contribution of Non-Financial Indicators. Final Year Project (Bachelor), Tunku Abdul Rahman University of Management and Technology.

Text RMM_Lim Sze Kei.pdf Restricted to Registered users only Download (1MB)

Abstract

Small and medium enterprises (SMEs) form an essential part of Malaysia's economy. Nonetheless, many such firms still struggle with accessing bank finance due to insufficient standardisation of the credit assessment process. This study explores the development and application of an MDA-based credit scoring model for Malaysian private SMEs and attempts to identify which non-financial variables add statistically significant incremental discriminant power over financial ratios. This study analyses an anonymised data set of 36 Malaysian SMEs, where six firms are labelled as financially distressed due to negative equity for three consecutive years. Six predictors were examined comprising three financial ratios (return on assets, current ratio, debt-to-asset ratio) and three non-financial indicators (company age, repayment behaviour, credit utilisation rate). Firstly, a financial baseline MDA model was constructed, followed by testing the incremental predictive ability of each of the non-financial predictors using partial F test. In the end, the model classifies the SMEs into three zones based on their risk profile. The result of testing shows that the baseline MDA model produces a statistically significant discriminant function, with Leverage being the most influential predictor among others, followed by liquidity and profitability measures. None of the non-financial variables reached 5% significance level. However, the reduction in Wilks' Lambda for Company Age suggests potential significance that may be constrained by the small sample size. The proposed risk classification framework successfully classified 83.3% of the SMEs. Financial ratios proved to be solid grounds for developing a credit scoring system based on the MDA method for SMEs. The model is practically applicable to pre-screening purposes. Future research should re-examine non-financial variables' predictive power using larger samples.