Logistic regression and multiple discriminant analysis in financial distress prediction : the case of selected industries in Malaysia

Nur Shafiqah, Tukino (2016) Logistic regression and multiple discriminant analysis in financial distress prediction : the case of selected industries in Malaysia. Masters thesis, Universiti Malaysia Sarawak.

PDF (Please get the password by email to repository@unimas.my, or call ext: 3914/ 3942/ 3933) Nur Shafiqah Tukino ft.pdf Restricted to Registered users only Download (4MB)

Abstract

In the past decades Malaysia recorded progress in its economic development. However, the figure of bankruptcies companies in Malaysia rose because our economic debt has doubled in the past five years. The need to ensure the health of companies in Malaysia is vital in order to reduce the unsecured debts or obligations within selected Malaysian companies such as construction sector, industrial product and consumer product. The samples of 270 companies of both distressed and non-distressed firms were used in this study during the period of 2004 until 2012. Thus, the Malaysian companies may prepare themselves to detect the financial flaws in their businesses. The focus point was merely to predict the financial distress using the Proposed Model with two methods: Multiple Discriminant Analysis and Logit Regression. Literally, most of the researchers declared that Logit Regression performed the best due to its accuracy level and often cited through decades. The results of the accuracy level using Multiple Discriminant Analysis (MDA) showed that in three-year period (t-3): 60% and one year period (t-1): 60% prior the event year increased the most compared to the accuracy level using Logit Regression in three-year period (t-3): 40% and one-year period (t-1): 53.3% in Proposed Model. Therefore, the study came to the point that Multiple Discriminant Analysis (MDA) contributed the most highest of accuracy level up to three years before the distress year compared to Logit Regression. Until now, the support in using Multiple Discriminant Analysis (MDA) is a good practice along with Logit Regression in forecasting the distressed level of the Malaysian companies as the investors or potential investors can benefit from these findings for better assess in the near future.

Actions (For repository members only: login required)

View Item