THE APPLICATION OF NEURAL NETWORKS AND A QUALITATIVE RESPONSE MODEL TO THE AUDITORS GOING-CONCERN UNCERTAINTY DECISION

An auditor gives a going concern uncertainty opinion when the client c ompany is at risk of failure or exhibits other signs of distress that threaten its ability to continue as a going concern. The decision to i ssue a going concern opinion is an unstructured task that requires the use of the auditor's judgment. In cases where judgment is required, t he auditor may benefit from the use of statistical analysis or other f orms of decision models to support the final decision. This study uses the generalized reduced gradient (GRG2) optimizer for neural network learning, a backpropagation neural network, and a legit model to predi ct which firms would receive audit reports reflecting a going concern uncertainty modification, The GRG2 optimizer has previously been used as a more efficient optimizer for solving business problems. The neura l network model formulated using GRG2 has the highest prediction accur acy of 95 percent. It performs best when tested with a small number of variables on a group of data sets, each containing 70 observations. W hile the legit procedure fails to converge when using our eight variab le model, the GRG2 based neural network analysis provides consistent r esults using either eight or four variable models. The GRG2 based neur al network is proposed as a robust alternative model for auditors to s upport their assessment of going concern uncertainty affecting the cli ent company.