Research on the financial risk measurement and prediction of small and medium-sized enterprises based on the artificial neural network under AHM-CRITIC coupling

Abstract

Based on AHM-CRITIC coupling weights combined with an artificial neural network to measure and study the financial risk of SMEs, the financial index system of four primary indicators and ten secondary indicators is constructed by collecting the publicly available financial report information of SMEs. The AHP matrix is obtained by the expert scoring method to calculate the subjective weights, and CRITIC calculates the objective weights, and the weights of both are coupled. The coupled weights are combined with the financial report data of LPMC to build a financial risk index model, and ANN neural network is used to predict and study the financial risk of the enterprise. The backtest curve fits the realistic data well. It can better predict the dramatic fluctuation of financial risk caused by the change of financial environment during the COVID-19 epidemic. The model is useful for SMEs to measure their own financial risks and make reasonable business decisions for the future by combining financial indicators.