Business Failure Prediction by Using the Hybrid Technique of GA and ANFIS Based on Financial Ratio: Evident from Listed Companies in the Stock Exchange of Thailand

Abstract

The objective of this study is to predict the business failure by using the combination of Genetic Algorithm (GA) and Adaptive Neuro Fuzzy Inference System (ANFIS). According to the forecasting procedure, 21 financial ratios are employed to use as the fundamental business failure signal for artificial intelligence generating within the combination technique. There are 109 Thai enterprises which are divided into 2 categories, for preparing the training set and test set. The process of this study starts with the GA selects attribute from the training set data group. After selecting process, ANFIS is taken to generate the forecasting model. Another set of data, test set, is used to prove the accuracy and error rate of testing result from the forecasting model. The result of the study points out that there are 5 financial ratios from GA-ANFIS selection which are appropriate to construct the failure forecasting model with 90.63 percent of accuracy rate and 3.13 percent of type 1 error. Finally, according to this study, the GA-ANFIS selection is helpful to predict the financial distress situation for Thai enterprises affirmatively.