Theoretical frontier and applied innovation of macro stress testing method based on internet and big data.
Since the end of the twentieth Century, along with the trend of financial globalization, financial derivatives have sprung up, financial industry transnational management gradually become the norm, the risk from a single institution can easily spread to a number of agencies, and even a national and international financial institutions. Especially after the subprime mortgage crisis in 2008, the financial stability has become the focus of attention in the theoretical and practical fields. Macro stress testing in the last century in 90s is considered to simulate the reaction in the financial system under the impact of extreme events, the International Monetary Fund, the World Bank and policy authorities conducted a lot of research and its application in the monitoring system of financial fragility in the practice of operation.
Financial data, refers to the use of big data to carry out financial services, namely for huge amounts of data, through the Internet, cloud computing and other information processing, the combination of traditional financial services, capital financing and financial service innovation. Big data finance through the platform finance and supply chain finance two models, the traditional financial model of mortgage loans to credit loans. This not only improves the financial efficiency, innovation of the financial model of the traditional financial industry, and the most important is to reconstruct the financial system, and promote cross-border integration of other industries. In more and more industries, dominated or enduring enterprises are often those who really can have big data, big data enterprise financial control.
At present, the financial institutions in many countries and government authorities have used the macro stress testing for risk assessment, but had not sorted and summarized systematically, which is not conducive to the understanding and study of macro stress testing, is not conducive to China's financial institutions to use the macro stress test monitoring risk. Therefore, it is necessary to systematically review and summarize the existing research in the past. The main contents of this paper are: (1) what is the concept of macro stress testing? What is the background? (2) what is the basic framework of macro stress testing? (3) the macro stress testing methods have what kind of practice and effect from the production to the present, and whether there are some development and improvement in practice? (4) what is the need to pay attention to the implementation of macro stress testing in our country? The above issues have always been the focus of the discussion of the scholars. This paper starts from the interpretation of the concept of macro stress testing method, introduces its background and development process, and discusses its theoretical framework, the research and practice of comparison, finally put forward the need to pay attention to China's macro stress testing problem.
2. Network big data and pressure test
2.1. Big data in the financial industry
Financial data refers to the massive collection of unstructured data, through big data, the Internet, cloud computing and other information, real-time analysis of customer consumption data, can provide customers a full range of information for financial enterprises, by analyzing and mining customer transactions and consumer information to grasp customer consumption habits, and predict customer behavior improve the financial service platform of new efficiency and reduce credit risk. Big data finance has the advantage of traditional financial difficult to compare, it can help enterprises closer to the customer, understanding customer needs, to achieve the non standardized exquisite service, increase customer stickiness. In the big data financial help, financial enterprises can also improve the credit system, to realize the innovation of credit management, reduce the rate of bad debts, to expand the scope of services, increase the proportion of the financing of Small and micro businesses, reduce the operating cost and service cost, so as to achieve economies of scale.
[FIGURE 1 OMITTED]
Through the big data mining, the financial services industry can use all the customer information to establish new relationships, dependence and correlation, thereby enhancing market competition ability, increase profits and income. Big data as a resource and tools, through in-depth mining will bring changes to the business model of the financial sector, providing the possibility of direct answers from other perspectives. At the same time, it may be possible to find new business opportunities and new business models. The so-called cloud computing is based on the increase of Internet related services, the use and delivery model, usually involving the Internet to provide dynamic and easy to expand and often virtual resources. Cloud computing for the importance of big data is reflected in the following aspects: massive data can be stored in the cloud server, according to the actual needs, in real time to expand or reduce the computing resources, etc. In short, the continuous development of cloud computing technology will reduce the cost of innovation in the big data mining industry.
[FIGURE 2 OMITTED]
The relationship between big data and Internet banking, on the one hand, the Internet banking as a new data source data; on the other hand, as a large data resources and tools that can be used for risk assessment and marketing of Internet banking, and even can be used for the Internet financial regulation. Different forms of Internet finance will produce large amounts of complex data, big data mining technology for Internet banking is an important tool, according to the Internet financial operation mode, from marketing, credit risk management to personalized financial services and other areas, big data mining technology can provide powerful support.
2.2. Stress test
Stress testing is a general term for all the techniques used by financial institutions to estimate the impact of extreme events on their robustness against some anomalies. Pressure testing applications in the micro field, there are two main roles. One is the Financial Soundness Indicators (such as CAMELS) to complement the risk measurement tool VaR, to assess the impact of some small probability events might cause banks or their own interbank assets portfolio; the two is to help the supervision of individual financial institutions risk of default. Therefore, after the Basel new capital agreement was announced, many developed countries regulatory authorities are required or encouraged by the bank to follow its recommendations to carry out the work of stress testing. Today, micro pressure test is widely regarded as a very important tool in risk management for the financial institutions of the banking industry in developed countries.
With the development of the financial environment, the periodic monitoring of a series of macro Prudential indicators is not enough to meet the needs of macro Prudential Management, and the development of quantitative tools for financial stability analysis has become an objective need. Macro stress testing is a major part of these quantitative tools.
Former world bank expert Sorge (2006) defined the system's stress test as an estimate of the risk exposure to financial institutions' special but reasonable stress situation. With different goals and micro pressure test on the personal portfolio of macro stress testing, the main purpose is to help the government identify the entire financial system may lead to systemic problems in the structure of vulnerabilities and exposures. With the outbreak of the financial crisis and crisis recovery in 2008 after the advance of macro stress testing is widely regarded as a new tool of effective management and solve the crisis, to guide bank restructuring, promote the recovery in confidence in the financial system role.
[FIGURE 3 OMITTED]
3. Development of financial stress tests
In the past twenty years, financial instability events occur frequently, had a huge impact, also attracted scholars and international financial institutions (such as the International Monetary Fund, the world bank and the Basel Committee) and the national regulatory authorities for the financial system (especially bank system) the stability of attention. Stress testing is an important and feasible method to measure the instability and vulnerability. Under the recommendation of the framework of the new Basel accord capital adequacy ratio, many large multinational banks in order to improve the internal control mechanism began a stress test. However, the risk of the entire financial system will be considered in the next few years by adding macro factors to the stress test.
In 1999, the International Monetary Fund (IMF) and the World Bank (Bank World) launched the financial sector assessment program (FSAP), stress testing for the first time as an important tool to measure financial vulnerability. In 2000, CGFS (Committee on the Global Financial System) to discuss the general pressure test (aggregate stress test) the potential use and interpretation of the concept of overall stress testing, stress testing is a general method of risk measurement of a group of financial institutions in the face of a specific scenario when exposed to .
Subsequently, the scholars will credit risk and market risk integration measurement, in the risk measurement to join the macroeconomic factors. D Worrell (2004) constructed a model of integrated macro stress testing, early warning system (warning system early) and financial vulnerability index. Allen (2004) put forward the credit risk measurement model to join the macroeconomic variables.
Figure 4--Stress testing framework What is stress testing * The need for stress testing Stress Testing Capital * Regulatory * Economic Methodology * Worst Case Loss * Shortfall methods Bank Capital Stress testing * Benchmanrs * Objectives Stress Testing Case * Banking * Insurance
Since the financial crisis of 2008, the FSAP stress test has been focused on the stability of the financial system. After constant revision and improvement, the FSAP stress test has become a true macro Prudential stress test. With the help of FSAP, member countries and regions have developed a pressure test system, there are more typical of the National Bank of Sweden DSGE development model, the development of the TD system, the Bank of England, the European Central Bank to establish the LGBGs model of the SRM system and the Central Bank of Austria. These systems are gradually added to the macro economic factors in the practice and testing of the system . According to Cihak (2006) statistics, as of the end of 25 there are more than 2005 countries or regions, the financial stability report released in the report relates to the content of the macro stress testing. In July 2007, the International Monetary Fund and the European Union and the European Central Bank experts and officials of the exchange in the "macro stress testing and Simulation of the financial crisis" meeting, they are all macro stress testing experience, put forward their own problems and discussed. As of 2015, the FSAP stress test was highly valued and widely welcomed by the 149 member countries and regions. At present, many countries and regions will be the FSAP pressure test as a necessary part of maintaining financial stability and promote financial development.
From the beginning of 2016, the International Monetary Fund (IMF) and World Bank (World Bank) of China, Germany, Britain, Russia, Sweden, Ireland, Finland, Belarus and other twenty countries to carry out a new round of financial sector two-year assessment program (FSAP) stress test.
4. The basic process of macro stress testing
According to whether the impact of the confined object is a single macro factor, the macro stress testing can be divided into two categories: sensitivity analysis and scenario analysis. The sensitivity analysis method is also called the single factor analysis method, is controlling for other macroeconomic factors (risk factor), a single impact of macroeconomic factors changes on bank asset portfolio and bearing capacity. The scenario analysis, also called the multi factor analysis, is to assume that a number of macroeconomic factors change the impact on bank assets or the entire financial system.
[FIGURE 5 OMITTED]
In comparison, the single factor analysis method is more simple, but there are many limitations in practice. One major advantage of this method is that the conduction process of various stress scenarios in the scenario analysis is relatively clear, which can allow multiple factors and changes in the same time.
[FIGURE 5 OMITTED]
1. The overall framework
According to the European Central Bank on 2013 issued Bank Solvency macro stress testing framework capability analysis (macro stress-testing framework for bank solvency analysis), the European Central Bank's macro stress testing framework is mainly composed of the design of macro financial situation, segmentation model analysis and evaluation of bank balance sheet changes and research feedback and contagion effects consist of four pillars.
[FIGURE 6 OMITTED]
The design of macro financial situation mainly carries on the design to make some of the banking system under pressure the impact of macro financial situation such as financing, the impact of macroeconomic shocks; pillar two (subdivision model analysis) will be a pillar of the scene into a reflection of the financial market and the impact of changes in bank balance sheet variables, and the establishment of credit risk and the interest rate risk, market risk, exchange rate risk and other different models; three pillars (assessment of balance sheet changes) is used after the breakdown of the model estimation, calculation of the profit and loss of assets and liabilities of banks, and the solvency of banks, and dynamic adjustment; four pillar from the scope of the entire financial system and real economy system. The banks and financial institutions after two times the effect of feedback and the contagion effect of solvency in the first time after impact Change.
2. Analysis of macro stress testing process
The macro pressure of national regulatory authorities and the banking industry testing process are similar, summed up can be seen, the key process of macro stress testing include: set the macro pressure and carry out risk stress testing, capital adequacy rate changes associated with the default response, the feedback effect of four modules
[FIGURE 7 OMITTED]
* Macro pressure scenario: Design macroeconomic scenarios generally have two steps: first, according to the stress test of the period, to identify the main systemic risk (such as exchange rate, price, etc.). Subsequently, the systemic risk factors into stress scenarios. The second step, using relevant models quantify these systemic risk factors the influence of the impact of financial institutions. Generally speaking, we can use the special event as the benchmark and model measurement as the basis of the two ways, the degree of variation of the risk factors, to achieve accurate and true reflection of the degree of risk factors by the macro pressure. After calibration, you can put the variables into the relevant dynamic macroeconomic model.
* Single risk stress test: That is, through the subdivision of the model, the assumption of pressure scenarios into independent risk variables, the use of specialized models to analyze the impact of different risks, the main risk of credit risk, interest rate risk and other market risks.
* Single agency test results summary: Through the balance sheet model, dynamic analysis method is used to simulate the changes of the bank balance sheet and the estimation of the solvency of the bank is obtained. The calculation of the profit and loss of the bank under compression includes four parts: the net interest income, the loan loss, the re evaluation of the market risk and the final calculation of the profit and loss of the bank. The solvency of the bank capital adequacy ratio index is usually used to measure, the final loss under the bank will increase, this part of the loss by bank capital to offset, in extreme adverse circumstances, risk weighted assets will increase.
* Correlation and feedback effects of institutional default: With the traditional "bottom-up" pressure test, macro stress testing framework of the impact on the real economy of the bank to adjust its liabilities, and to measure the two impacts of bank assets and capital adequacy ratio of the deterioration in the real economy, namely the feedback effect. At the same time, the macro stress tests will be conducted for the financial institution's default correlation analysis.
[FIGURE 8 OMITTED]
In infectious analysis, assuming that the financial institutions Di default, then through to hold [D.sub.1] risk assets (such as stored in the [D.sub.1] deposit, interbank lending, financial bonds, long trading positions, reverse repurchase assets etc.) financial institutions ([D.sub.2], [D.sub.3]) survey, estimates will lead to [D.sub.2] [D.sub.3] ... And other financial institutions, the capital adequacy ratio fell below the regulatory limit. In addition to the balance sheet items, but also pay attention to bank acceptance bills of exchange, security class business, financial derivatives trading and other tables outside the business.
5. Research and practice of Macro stress testing
The theoretical research of macro stress testing mainly focuses on the construction of the framework. Sorge (2004) compared the basic principles and methods of bank stress testing. Stolz (2008) introduces the difference and connection between micro pressure test and macro stress testing, and introduces the framework and process of the International Monetary Fund's macro stress testing. A large number of scholars have summed up their respective countries (regions) to construct the macro stress testing framework, such as Ross (2006), Ryan (2007), Lind (2007), Krimminger (2008), Wong (2009), Rongjie (2011), Catalin Ruja (2014). Since 2007, the inter-bank market feedback effect (Effects Feedback), Effectsor Domino (Contagion Effects) research gradually increased.
The empirical field of macro stress testing is mainly focused on finding the probability of default macro decision variables and the degree of the influence, find the relationship and law of risk indicators and macroeconomic shocks among the variables, in order to achieve the purpose of measuring the risk of macro stress testing. Blaschke (2001) pointed out that the selection of a number of banks to analyze the interbank market system risk. Boss (2007) in the credit model to join the macroeconomic variables, the design of the pressure scenario of the Australian banking system, through the calculation of the probability of default of a single industry, plus the overall probability of default. Boss using data from the Australian banking industry, the results show that the gross industrial product, inflation, stock index and oil prices have a significant impact on the probability of default four. Goodhart (20042007), Tsatsaronis (2007) and Summer (2007) are concerned with the problem of endogenous risk and nonlinear effect in the derivatives market. Miroslav and David (2008) pointed out that the impact of extreme situation, the linear measurement equation can accurately quantify the macroeconomic fluctuation, unable to accurately measure the financial risk, on this basis, Miroslav and David to construct high order nonlinear measurement equation based on the interpretation of variables, and achieve a good effect. Et al. Huang (2009) in order to spread the bank in order to spread the risk of insurance premiums as a risk indicator, in the design of the pressure scenario to consider the probability of default and the correlation between bank assets. Credit default swap (CDS) market and stock market data are used to calculate the probability of default and the risk of bank assets. Havrylchyk (2010) using the logit model of loan loss reserve and macroeconomic variables to establish contact, starting from the pressure setting, studied under the impact of macroeconomic variables, how banking loan losses will change. Coffinet and Lin (2010) through the establishment of the French bank's profitability (ROA) and the French macroeconomic indicators, simulation when the macroeconomic downturn, the changes in the profitability of each bank. Ali (2010) by using the rate of bad loans as a risk index, designed the same pressure shock scenario, in this case the calculation of the United States and Australia non-performing loan ratio, the conclusion is under the same macroeconomic variables, the performance in Australia is better than the United States.
In the field of practice, this article describes the FSAP pressure test system, based on the FSAP pressure test system, and then introduces the more mature TD and SRM system:
1. FSAP pressure test system
Since the end of the Asian financial crisis in 90s, the International Monetary Fund and the world bank have summed up the experience and lessons of the crisis. In May 1999, IMF and World Bank jointly launched the "financial sector assessment program" (Financial Sector Assessment Program, FSAP), to member countries and the economies of other banks and the entire financial system of comprehensive assessment and monitoring. Over the past twenty years, through continuous development and improvement, FSAP as a financial stability assessment of the commonly used framework has been widely accepted in the world. Under the framework of the financial sector assessment framework for financial stability assessment, usually there are financial stability indicators, stress tests and standards and guidelines for evaluation of three commonly used analytical tools. After the financial crisis in 2008, FSAP further considered the impact of macroeconomic factors on the stability of the entire financial system, built a sound macro Prudential stress testing system.
FSAP pressure testing system has become the basic framework for national development pressure test system, which mainly includes six modules: credit risk, interest rate risk, exchange rate risk, liquidity risk, operational risk and risk contagion. Risk scenarios, the impact of indicators of the design are based on each country's macroeconomic policy, the environment, whether the data can be a problem, and some differences. In general, in most countries the use of non-performing loan ratio (NPL) analysis of credit risk, the complexity of some of the selected default probability (PDs) or default loss rate (LGDs). But some countries choose foreign currency loans (such as Armenia), some countries have chosen cross-border lending (such as Austria), and the country chose to use the loan concentration (such as Russia) to measure credit risk. The measure of exchange rate risk, some countries choose to use the net position of some countries (such as Sweden), the capital at risk (such as France and Germany); contagion among banks choose unsecured loans, property, payment system and other positions to measure risk.
Today's FSAP pay more attention to macroeconomic factors in the design of scene, and they were testing countries (or regions) close cooperation policy authorities and the bank, the micro pressure test from the bottom to the top of macro stress tests and from top to bottom combined to expand the scope of non banks, interbank contagion effect considered and in the test.
2. TD pressure test system of Bank of England
In 2006 the England bank issued a financial stability report (FSR, 2006) before the British financial system mainly adopts pressure test single factor analysis method of stability, without considering the interaction between banks, contagion effect, also do not consider the macroeconomic and financial institutions between the feedback effect, enterprises and financial institutions in the financial accelerator effect etc.. In 2006, the Bank of England began to use the new pressure test system to assess the risk of financial stability. In the financial policy committee (FPC) on the advice and cooperation, in October 2013, the Bank of England issued a draft regular pressure test on the UK banking system, and the stress testing framework to the relevant parties for comments. Subsequently, the parties to listen to the feedback on the basis of the Bank of England in 2014, 2015 conducted a two stress test.
Unlike other stress testing systems, the Bank of England's TD stress testing system, in addition to the usual risk studies, is also taking into account the liquidity risk. TD system stressed from top to bottom, the vulnerability of the financial system at the core of the financial sector and the real economy, the risk of infection between the channels of the channel has been the focus of attention. The system of the default rate causes the bank credit rating downgrade, endogeneity, rising financing costs and other risks of the contagion effect, inter agency interaction and the risk of financial institutions and macroeconomic feedback effects etc.. The pressure test also draws on the concept of system dynamics, taking into account the amplification mechanism and the spillover effect.
3. The SRM system of the Central Bank of Austria
In 2002, Oesterreichische Nationalbank, the financial market authority (Austrian Financial Market Authority, FMA) and some scholars jointly formulated the system of risk monitoring system (Systemic Risk Monistor, SRM). This system is mainly based on the Boss (2002) and Elsinger (2006) of the interbank network model, using the inter-bank market to reflect the relationship between bank lending positions. The model is divided into two parts, on the one hand, the use of a single bank's assets and liabilities, to predict the future may face the market risk and credit risk and loss distribution; on the other hand, the mutual position survey of interbank positions, to build the bank network model according to the situation, assess the market risk of default. Using the macro-economic history data, the joint distribution of various risks, such as credit risk, interest rate risk, exchange rate risk, and so on, is constructed. Based on the Monte Carlo simulation, the joint distribution of the macro economic variables is obtained. To a certain extent, the model considers the transmission mechanism, but the feedback effect is not considered.
6. Conclusion and suggestion
China is gradually joining the global economy, the trend of financial globalization has made China face a complex and volatile international financial environment. Our country is deepening financial reform crucial stage and the deep water, because of the particularity of our country economy, political system, social stage, building and construction, many of the implementation of the system and improvement of many policy measures have no experience can learn. Therefore, we need to use macro stress testing to predict the macroeconomic changes in the macroeconomic policy changes, will affect the stability of the financial system, so as to improve our financial policy, reduce risk.
With the slowdown in global economic growth, China's economic soft landing, banks and the entire financial system will have a drastic change. The macro stress tests have a key role in identifying, preventing and dissolving the risk of banking and the whole financial system. However, in the practice of our country, the application of macro stress testing methods to assess the level of the stability of the banking system, there are still many challenges. For example, how to optimize the test flow, how to break the limitations of the data, whether the scope of the test should be extended, how to apply the results of the macro stress tests, etc.
Through summarizing the research literature and drawing lessons from practical experience, this paper believes that the future of China's macro stress testing needs to pay attention to the following three aspects.
1. First, the design of effective macro stress testing objectives, to improve the accuracy of the pressure test. On the one hand, the introduction of a single financial institutions in general equilibrium. On the other hand, pay attention to the whole system of financial flexibility test. FSAP embedded in the financial stability framework, will be a combination of macro Prudential stress testing and micro Prudential stress testing, while carrying out the bottom-up and top- down stress tests.
2. Second, to expand the impact coverage, considering the risk of cross-border contagion and other financial institutions, the impact of the real economy sector on macroeconomic variables. To consider the impact of other institutions in the financial markets, such as the impact on the initial period and the impact of the impact of the shadow banking, as well as trust companies, money market funds, P2P platform, private finance companies, etc.. With cross-border business has become the norm, international trade, global economic integration, a country affected by macroeconomic shocks will rapidly spread to other countries or regions.
3. Third, build and improve the index system, improve the level of early warning. Establish and gradually improve the Financial Soundness Indicators and early risk warning system. Secondly, select a reasonable indicator system. Through the practice of the countries and scholars of the theoretical study of the review, it can be seen that the choice of reasonable indicators is particularly important. Can refer to foreign experience, select the macro economic indicators with Chinese characteristics to improve the risk impact index system, such as the entrepreneur confidence index, the national housing boom index.
Allen, L., Saunders, A. (2004). Incorporating Systemic Influences Into Risk Measurements: A Survey of the Literature. Journal of Financial Services Research, 26(2), 161-191.
Boss, M. (2002). A macroeconomic credit risk model for stress testing the Austrian credit portfolio. Financial stability report, 4, 64-82.
Blaschke, W, Jones, M. (2001). Stress Testing of Financial Systems: An Overview of Issues, Methodologies, and FSAP Experiences. Imf Working Papers, 01, 1-56.
Coffinet, J., Lin, S. (2009). Stress testing French banks' income subcomponents. Ssrn Electronic Journal, 242-250.
Catalin, R. (2014). Macro Stress-Testing Credit Risk in Romanian Banking System (Dissertation thesis). Mpra Paper, 20-24.
Chen, J., Joe Z. (2011). Characteristics of macro stress testing, implementation framework and policy recommendations. Financial forum, (7), 32-36.
Drehmann, M. (2008). Stress Tests: Objectives, Reality, Limitations and Possibilities. ECB, 54-56.
Delgado, A., Velthuis, M. (2015). Proposal for a continuous improvement IT governance framework at financial institutions. RISTI--Revista Iberica de Sistemas e Tecnologias de Informacao, (15), 51-67.
Guo, F., Fung, H. (2009). The Dynamic Impact of Macro Shocks on Insurance Premiums. Journal of Financial Services Research, 35(3), 225-244.
Havrylchyk, O. (2010). A macroeconomic credit risk model for stress testing the South African banking sector. Mpra Paper, 20-25.
Haldane, A., Hall, S., Pezzini, S. (2007). Financial Stability Paper No 2: A New Approach to Assessing Risks to Financial Stability. Applied Physics B, 377(1), 101-108.
Krimminger, M. (2008). The resolution of cross-border banks: Issues for deposit insurers and proposals for cooperation. Journal of Financial Stability, 4(4), 376-390.
Rongjie, T, Yang, J. (2011). Macro Stress Testing on Credit Risk of Commercial Banks in China Based on Vector Autoregression Models. Ssrn Electronic Journal, 11-15.
Sorge, M, Virolainen, K. (2006). A comparative analysis of macro stress-testing methodologies with application to Finland. Journal of Financial Stability, 2(2), 113-151.
Sorge, M. (2004). Stress-testing Financial Systems: An Overview of Current Methodologies. Ssrn Electronic Journal, 68(3), 18-21.
Stolz, S., Moretti, M., Swinburne, M. (2008). Stress Testing at the IMF. Social Science Electronic Publishing, 08, 1-23.
Worrell, D. L. (2004). Quantitative Assessment of the Financial Sector: An Integrated Approach. Social Science Electronic Publishing, 04, 150-153.
Wong, J, Choi, K. F. (2009). A framework for macro stress testing the credit risk of banks in Hong Kong. Journal of Financial Stability, 20-29.
Zhang, L. (2016). The global effectiveness of the macro Prudential stress test. China finance, 70-75.
Lei Yang, Qiang Zhang
Hunan University, Changsha 410007, China
|Printer friendly Cite/link Email Feedback|
|Author:||Yang, Lei; Zhang, Qiang|
|Publication:||RISTI (Revista Iberica de Sistemas e Tecnologias de Informacao)|
|Date:||Dec 1, 2016|
|Previous Article:||Research on spatial distribution of tourist attractions and service facilities scale planning based on GIS.|
|Next Article:||Data mining algorithm application in influencing factors analysis of college students' physical health.|