Listed Company Financial Crisis Early Warning in the Electronic Information Industry Based on QBPSO-SVM

The purpose of this research paper is to construct a financial crisis early warning model for Chinese listed companies in the electronic information industry, validate the model based on the collected data, and give corresponding suggestions based on the experimental results. The main research objectives of this paper are 1) to identify 13 financial indicators and crisis factors; 2) to establish an early warning model and indicator system; and 3) to validate and improve the early warning model. The research method of this paper is the quantitative study of 13 financial indicators combined with the algorithm of Quantum Behavioral Particle Swarm Optimization Support Vector Machine to verify the applicability of the early warning model of financial crisis for listed companies in the electronic information industry. The research content is through the China Securities Regulatory Commission designated information disclosure website of listed companies CNINFO (www.cninfo.com.cn) to select the economy of the top 155 A-share listed companies in the electronic information industry as the research sample overall, according to the principle of financial indicators to determine the final selection of the sample group of 76 A-share listed companies in the electronic information industry. Considering the variability of market data to determine the screening statement data for 2017–2021. The designated samples must comply with the domestic accounting standards; no bankruptcy, no write-off of normal operation of the company, from the electronic information industry, with the shape of the asset size, the same time period. Experimental validation of quantum behavior particle swarm optimization support vector machine financial crisis early warning model for electronic information industry is carried out through matlab modeling software. The specific algorithm is to optimize the SVM by QBPSO algorithm to get the optimal parameters, import the financial index data from 2017 to 2021 into the model test, and the final result reaches 94.54%. The results of the research are as follows: this thesis uses the quantum behavior particle swarm optimization algorithm on the utility of support vector machine in the financial crisis early warning of listed companies in the electronic information industry.