A Study On Risk Early-warning System Of Security Companies

The core of modern economy is the finance, and finance is the core of the capital market. Securities companies, as the main participants in capital market, is an important driving force and the engine in the capital market. As the achievement of the functions of capital market lies on the intermediary role of a bridge of securities companies, so the importance of securities companies is self-evident.Risk is the inherent characteristics of securities companies. Being the industry's high-risk characteristics, the survival of securities companies is based on the ability of risk prevention and control, and the basic conditions for its development is also due to strict and effective risk management. Internationally, since the 80's, the international cyclical financial market turmoil has been intensified continuously. In 2007,the outbreak of the loan crisis of the United States has transmited and evoluted into the financial crisis that happened once within one century, and has led to economic crisis and the global economic recession. Its fuse is from over-leveraged and over-flooding sub-loans in the secondary derivative financial products from those so-called global Securities compnies such as AIG, Lehman Brothers, Merrill Lynch and so on. Its potential risks go far beyond the affordability of a single securities company, the securities industry or even the financial system, and ultimately resulted in catastrophic consequences. In the course of the development of domestic capital markets in two decades, risk management of securities companies went out of control, which brought about numerous illegal issues. These facts show that it is very important to build a comprehensive, effective, dynamic and forward-looking risk early-warning system of securities firms.At present, resulting from the limitations of research ideas, research aspects, and research methods, the traditional early-warning system of securities companies is limited to micro-quantifiable risk identification, measurement and control of internal securities companies, and is lack of a combination of external qualitative risks. So, it can not fully and accurately measure the risk of securities firms, and aslo can not meet the need of macro-risk early-warning management for the industry. In an increasingly complex economic and financial environment, the risk early-warning system of securities companies, is not only required to be applicable to securities firms to achieve a comprehensive and accurate identification, measurement, assessment and management of internal and external risks, but should also to apply to industry regulatory authorities to achieve industry macro-risk early-warning management. Innovation requirements. To prevent a single financial product, the risk of a single securities company growing into the industry systemic risk. Firstly, on the basis of summing up the related research results at home and abroad, the paper proposes the general structure modules and operation process of risk early-warning system of the securities companies. and the structure is composed of five modules such as Risk Recognition, Risk Measurement, Risk Comprehensive Judgment, Early-warning Signal Output and System Evaluation and Amendment. In addtion, related discussions of theory and methodology about the construction, functions, and operation process of the system structure modules are explored in this paper to build the model framework for system construction.Secondly, according to different construction and handling approacches of risk early-warning indicator system during system construction, as well as different perspectives of early-warning generating signals methods, this paper establishes two types of application models of risk early-warning system of securities companies - signal discriminance applications model based on Logistic regression modeling, and direct quantitative signal extraction application model;Then, based on the use purpose and requirements of different users levels of early-warning system (securities companies, industry regulators, etc.),this paper proposes the principles of preventive, timeliness, operability, universality and effectiveness of system construction, as the standards of judgement and evaluation of two system application models.In accordance with these standards, to judge the effectiveness of the Logistic regression method application model, this paper conducts the model regression by using samples data in 2003 from 19 securities companies in Guangdong Province with the most problems during the bear market, as well as 60 national securities companies. and the result shows that early-warning signal of early-warning system application model of Logistic regression method has some validity, while there are some application limitations of the model in terms of too narrow range of options of precaution indicator system and the lackness of forward-looking of early-warning signals, which can not satisfy the principles of preventive and effectiveness of system construction.In order to effectively build the system application model in accordance with above principles, this paper further sorts out the limitations and weaknesses of risk early-warning system of securities companies in the past study, and raises the ideas and princibles of system construction with structure innovation, methodological innovation and the application innovation, and achieve the combination between external macro-risks and internal micro-risks, and achieve the combination of two levels between securities companies and regulatory institutions. Moreover, using the innovative system Metasynthesis integration method from qualitative to quantitative, this paper completes the innovative construction process of risk early-warning system of securities companies based on the signal extraction method.To judge the innovative model of risk early-warning system, this paper carries out the simulation of early warning both from two user levels of securities firms and regulatory bodies, that is from two ways of internal control and macro-management. For simulation of early warning of the internal control of securities companies, this paper conducts simulation of application of 20 randomly selected securities companies to verify the timeliness and effectiveness of the system. On the other hand, for the simulation of early warning of regulatory bodies'macro-management, this paper starts with the application of simulation of early warning through the vertical and horizontal dimensions. And for the vertical simulation of early warning, this paper is through simulation applications of 80 securities companies at the same time, to verify the overall effectiveness of the system. While for horizontal simulation of early warning, by removing the closed or under-trusteeship companies during the sampling period, this paper conducts the horizontal simulation of application of the comparable 60 securities companies for 5 consecutive years from year 2003 to year 2007 to verify the dynamic and forward-looking of the system. Ultimately, this paper comes to the evaluation findings of good effect of early-warning signals, and the system can achieve the internal control requirements of securities companies and the macro early-warning management of regulatory bodies too.Finally, this paper proposes related policy recommendations of risk prevention and control policy of securities companies.