Analysis of the environmental trend of network finance and its influence on traditional commercial banks

https://doi.org/10.1016/j.cam.2020.112907Get rights and content

Abstract

In order to understand the environmental trend of network finance and the impact of network finance on traditional ommercial banks, after analyzing the transaction risks of network finance, this study learns from the advanced models of credit risk measurement and early warning at home and abroad and the latest artificial intelligence technology, combining them with China’s national conditions, so as to establish a credit risk measurement system suitable for commercial banks. In this study, the research is carried out from the perspectives of theory and practice, technology and business, and the credit risk warning system, measurement model and implementation tools are comprehensively sorted out, then the basic theories and core ideas are studied. In accordance with the concept of big data mining, this study proposes a financial crisis early warning model based on artificial intelligence system. Based on the empirical analysis of the main causes of the rapid decline in the asset quality of China’s joint-stock​ commercial banks, and based on the characteristics of big data mining in the information explosion era, it is pointed out that the artificial intelligence is a powerful tool to improve the ability of credit risk measurement. It can be confirmed that the development of network finance has brought different impacts on the business, business model, and business philosophy of banks.

Introduction

The financial crisis has hit the international banking industry hard and exposed serious problems in credit risk management of commercial banks. Finance plays a very important role in the process of economic development [1]. Since the 21st century, the financial market has continuously created miracles to help the world economy maintain rapid development [2], [3], [4]. Especially in the new century, western economies represented by the United States have achieved sustained and rapid growth for 8 consecutive years, which has also driven the explosive growth of the entire financial industry, accompanied by the continuous innovation of various financial derivatives [5]. However, no matter how financial products evolve, the essence of credit risk still exists. Neglecting the control of credit risk will inevitably bring about a huge risk outbreak. Behind the sustained economic prosperity, potential risks are gradually accumulating, the exposure to risks is constantly enlarging under the leverage of many financial derivatives, and the threat is also multiplying [6]. Affected by the US subprime mortgage crisis in 2008, a global financial crisis broke out, which had a significant negative impact on the global economy. The crisis lasted for many years. The US financial giants Fannie Mae and Freddie Mac were taken over by the government. Lehman Brothers, AIG, Merrill Lynch and HBOS successively declared bankruptcy. Many important financial institutions were on the verge of bankruptcy and the financial industry was in mourning [7], [8], [9]. This time, global financial institutions have been hit hard and have had a great impact on the global economy. Banks in developed countries have also been hit hard. The quality of assets has been deteriorating, loan losses have been rising, capital adequacy ratio has been seriously eroded, and credit risk has been significantly increased, which has had a great impact on the real economy. There are many reasons for the crisis, but one of the key ones is that credit risk management is not in place, which makes people begin to rethink the credit risk management of business [10], [11]. Credit risk is the biggest risk faced by commercial banks in their operations, and it is also the source of commercial banks’ operating income. How to effectively control credit risk will directly affect the operation of commercial banks.

With the increasing number of Internet financial products and the deepening of Internet enterprises’ exploration of financial business, it has a huge impact on China’s financial industry. This also requires traditional commercial banks to carefully evaluate the impact of the development of Internet finance on themselves, as the basis of innovative business and service types [12]. At the same time, the financial supervision department should effectively supervise and manage the Internet finance to ensure the stability and order of the financial sector. In the Internet era, the impact of the rapid development of Internet Finance on the business operation of traditional commercial banks is analyzed. Then, it is put forward that in the face of the strong impact of Internet finance, how commercial banks face the crisis is an urgent problem to be solved.

In this study, advanced mature credit risk measurement models and prediction technologies at home and abroad are thoroughly studied and summarized, and their basic theories and core ideas are studied. On the basis of empirical analysis of the main reasons for the rapid decline in asset quality of joint-stock commercial banks in China, the credit risk early warning model of commercial banks is analyzed and studied. Through the analysis and comparison of credit risk measurement tools, according to the complexity of credit risk and the reality of information technology system construction in domestic commercial banks, the necessity for commercial banks to introduce advanced technologies such as artificial intelligence for credit risk measurement analysis is proposed [13], [14], [15]. On the basis of in-depth research on the structure and algorithm of the deep confidence network, and aiming at the characteristics of supervised learning classification algorithm, an optimization improvement method of restricted Boltzmann machine algorithm with supervised learning classification problem is proposed [16]. The performance is verified by manually generating data, which shows that the algorithm is superior to the algorithm before optimization. In this study, an enterprise’s single-family financial crisis early warning and financial index prediction model is established. Based on the complete sample data of listed companies from 1993 to 2015, an empirical study has been carried out and good results have been achieved. The research results of this study not only have solid theoretical basis, but also have strong operability and wide applicability, which can provide risk warning ideas for senior managers of commercial banks. This study makes a beneficial attempt from the integration of theory, business, technology and other dimensions [17], [18], [19]. Commercial bank system developers can simply process and adjust according to their own data according to the model in this study, and directly build their own risk warning system after personalized transformation, which can be applied to the bank’s management practice. At the same time, it can also provide reference for regulatory authorities, guide commercial banks to strengthen the research on frontier theory and technology, improve the existing credit risk system, enhance the data mining ability, effectively improve the credit risk management level, and ultimately enhance the international competitiveness of China’s commercial banks.

Section snippets

The influence of internet finance on traditional commercial banks

At present, the main source of profit of traditional commercial banks in China is credit business, and the income of credit business mainly comes from interest margin income. The rapid development of Internet finance has begun to threaten the interest margin income of commercial banks [20]. The development of Internet lending mode meets the financial needs of small customers who are not valued by commercial banks before. Using big data technology, the Internet financial platform can discover

Deep confidence network

Neural network is a network model built to simulate human brain. Human brain is a multi-level network, but previous studies found that the learning effect of deep network is not as good as that of shallow network. In 2006, Hinton introduced unsupervised feature learning to effectively solve this problem. Deep learning expresses complex representations through other simpler representations, which solves the core problems in representation learning. Deep learning is the development of neural

Hardware and software environment

Because the analysis and prediction of financial data need a lot of data, there are usually many ways to obtain the stock data. TuShare is a free financial data interface package, which mainly uses Python language to obtain most of the financial market data such as stocks, funds, futures, etc. TuShare interface not only needs the support of transaction data over the years, but also needs multi-dimensional analysis of basic data. As an important part of financial data, stock data is arranged in

Conclusion

In this study, the advantages of deep learning in financial data analysis are introduced. Therefore, in this study, deep learning algorithm is applied to financial data analysis system, the main functions include data collection, and data analysis and so on. This study proposes a Boltzmann machine learning method with restricted classification and partition on the basis of in-depth confidence network learning algorithm, and select all data of A-share listed companies for empirical research. The

Acknowledgment

It was supported by the National Natural Science Foundation of China (Grant No. 71850006).

References (34)

  • SirignanoJ. et al.

    Universal features of price formation in financial markets: perspectives from deep learning

    Quant. Finance

    (2019)
  • SunT. et al.

    Embracing textual data analytics in auditing with deep learning

    Int. J. Digit. Account. Res.

    (2018)
  • CulkinR. et al.

    Machine learning in finance: the case of deep learning for option pricing

    J. Invest. Manage.

    (2017)
  • AggarwalS. et al.

    Deep investment in financial markets using deep learning models

    Int. J. Comput. Appl.

    (2017)
  • LiangY. et al.

    Deep learning based inference of private information using embedded sensors in smart devices

    IEEE Netw.

    (2018)
  • DingliA. et al.

    Financial time series forecasting–a deep learning approach

    Int. J. Mach. Learn. Comput.

    (2017)
  • PérilleuxA. et al.

    Are financial cooperatives crowded out by commercial banks in the process of financial sector development

    Kyklos

    (2016)
  • Cited by (9)

    • Does digital finance aggravate bank competition? Evidence from China

      2023, Research in International Business and Finance
    • Research on credit rating and risk measurement of electricity retailers based on Bayesian Best Worst Method-Cloud Model and improved Credit Metrics model in China's power market

      2022, Energy
      Citation Excerpt :

      Literature [27] propose a robust cost-sensitive kernel method with Blinex loss (CSKB), which can be applied in credit risk evaluation. Literature [28] learns from the advanced models of credit risk measurement and early warning at home and abroad and the latest artificial intelligence technology, so as to establish a credit risk measurement system suitable for commercial banks. Literature [29] survey existing researches and methodologies on assessment and measurement of financial systemic risk combined with machine learning technologies, including big data analysis, network analysis and sentiment analysis.

    • Revisiting financial development and renewable energy electricity role in attaining China's carbon neutrality target

      2021, Journal of Environmental Management
      Citation Excerpt :

      However, to finance these environmentally friendly projects, a well-developed financial system is a pre-requisite (Ali et al., 2014; Ji et al., 2020). However, China's financial system has undergone by leaps and bounds, but it is still dominated by state-owned banks (Xu et al., 2020). The high concentration ratio and dominance of the financial system by major banks lead to a low degree of competition within the financial institutions.

    View all citing articles on Scopus
    View full text