CN107609784A - Utilize the system and method for big data quantitative analysis foreign exchange investment risk - Google Patents
Utilize the system and method for big data quantitative analysis foreign exchange investment risk Download PDFInfo
- Publication number
- CN107609784A CN107609784A CN201710875808.8A CN201710875808A CN107609784A CN 107609784 A CN107609784 A CN 107609784A CN 201710875808 A CN201710875808 A CN 201710875808A CN 107609784 A CN107609784 A CN 107609784A
- Authority
- CN
- China
- Prior art keywords
- data
- foreign exchange
- risk
- model
- big data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The invention belongs to big data applied technical field, it discloses a kind of system and method using big data quantitative analysis foreign exchange investment risk, to assess greatest risk penalty values of the different delivery time difference foreign exchanges under the fluctuation of market normal dynamic, visual reference is carried out for enterprise.This method includes：A. by bottom foreign currency operation system data, third party website data and part manual data, database purchase is introduced by different modes；B. the data demand that foreign currency operation data propose according to model carries out ETL processing；C. the initial data to collection and the data of data processing were carried out it is associated storage；D. model foundation is carried out according to reduced data, by continuing to optimize and machine learning, final output big data foreign exchange investment VaR risk models measure to quantifying riskization；E. the measurement result of foreign exchange investment quantifying risk model is shown in a manner of graph visualization.
Description
Technical field
The invention belongs to big data applied technical field, and in particular to one kind utilizes big data quantitative analysis foreign exchange investment wind
The system and method for danger.
Background technology
Financial risk management is content most crucial in whole business that all kinds of financial institutions are engaged in and management activity, it
With the time value, asset pricing by and be known as three big pillars of modern finance theory.Financial risk management be divided into identification risk,
Measure four risk, the assessment of processing risk and risk management and adjustment steps.Wherein, the measurement of financial risks is financial city
The core link of field risk management.The quality of risk measurement, has been largely fixed the validity of financial market risks management；
The selection of reasonable Risk Measurement index, it is the effective guarantee for improving risk measurement quality.Therefore with Finance and mathematical statistics
Theory combines a kind of various market risks of asset portfolio with a single index (VaR values) to weigh, and structure is big
Data foreign exchange investment VaR risk models arise at the historic moment.
Traditional exchange risk measuring and calculating is with the fluctuation method that variance and risks and assumptions etc. are main Measure Indexes, this side
Accurately estimation of the method to asset portfolio future profits rate probability distribution is relatively difficult, it is common to use just too distribution often deviate from reality
Border；Fluctuation method merely depict the degree of fluctuation of asset portfolio future profits, can not illustrate that portfolio value changes
Direction；Fluctuation method can not provide the concrete numerical value of portfolio value change.
VaR English full name is Value at Risk, and it refers to the part being exposed in assets value in risk, can claim
For venture worth.Refer under the conditions of normal market fluctuation and in given delivery time, a certain foreign exchange funds is following specific
Maximum possible loss value in one section of holding period.Quantify a kind of foreign exchange and hold the risk exposure journey completed a business transaction to future market fluctuation
Degree, is commonly available to financial environment and compares the flation.By foreign exchange investment quantifying risk.
The difficulty of foreign exchange investment risk model based on big data is to determine different delivery time difference foreign exchanges in city
Downside Risk under the normal dynamic fluctuation of field, can not determine every foreign exchange transaction optimal delivery time.
The content of the invention
The technical problems to be solved by the invention are：It is proposed a kind of be using big data quantitative analysis foreign exchange investment risk
System and method, to assess greatest risk penalty values of the different delivery time difference foreign exchanges under the fluctuation of market normal dynamic, for enterprise
Industry carries out visual reference.
The present invention solves the technical scheme that above-mentioned technical problem uses：
Using the system of big data quantitative analysis foreign exchange investment risk, including：Data acquisition module, data memory module,
Data processing module, big data foreign exchange investment VaR risk models algoritic module, model result output display module；
The data acquisition module, it is that bottom foreign currency operation system data, third party website data and part is manual
Data, database purchase is introduced by different modes；
The data memory module is to the initial data of collection and carried out the data of data processing and is associated and deposits
Storage, and support increment to store；
The data processing module, it is that the data demand proposed to foreign exchange business datum according to model carries out ETL processing；
The big data foreign exchange investment VaR risk model algoritic modules, for carrying out model foundation according to reduced data,
By continuing to optimize and machine learning, final output big data foreign exchange investment VaR risk models measure to quantifying riskization；
The model result exports display module, for by the measurement result of foreign exchange investment quantifying risk model with figure
Visual means are shown.
As further optimization, the measurement result of the foreign exchange investment quantifying risk model includes：Difference is calculated to complete a business transaction
The maximum loss value of phase difference foreign exchange, and suggest every foreign exchange transaction optimal delivery time.
In addition, the invention also provides a kind of method using big data quantitative analysis foreign exchange investment risk, it include with
Lower step：
A. by bottom foreign currency operation system data, third party website data and part manual data, different sides is passed through
Formula introduces database purchase；
B. the data demand that foreign currency operation data propose according to model carries out ETL processing；
C. the initial data to collection and the data of data processing were carried out it is associated storage；
D. model foundation is carried out according to reduced data, by continuing to optimize and machine learning, outside final output big data
The investment VaR risk models that converge measure to quantifying riskization；
E. the measurement result of foreign exchange investment quantifying risk model is shown in a manner of graph visualization.
As further optimization, step a is specifically included：
Prepare data-interface access system internal user information, accounts information, sequence information, procurement information and outside country
Foreign exchange office data, then it is stored in database mysql；
Reptile obtains foreign exchange market industry and financial policy dynamic data, and extra festivals or holidays data deposit database
mysql；
Third party's data excel is imported in database mysql by java program forms.
Optimize as further, in step b, the ETL processing includes：Data needed for model are pre-processed to be applicable
Used in model, specifically include data normalization processing and the processing of shortage of data value.
Optimize as further, in step c, the initial data of the collection includes structuring and non-structured data.
As further optimization, step d is specifically included：
D1, index system are established：According to the factor for influenceing big data foreign exchange investment VaR risk models, service index system
The feature selecting flow of foundation carries out the screening and combination of index feature, and output is applied to big data foreign exchange investment VaR risk moulds
The index feature system of type；
D2, model learning：Model obtained from database mysql user profile, accounts information, sequence information, procurement information,
Foreign exchange currency type, foreign exchange rate, systematic parameter, carry out big data foreign exchange investment VaR risk models and calculate, and by result input data
In the mysql of storehouse；
D3, model optimization：By being continued to optimize to index, the continuous training to model, final big data foreign exchange is exported
Invest VaR risk models.
Optimize as further, in step e, in addition to：Different friendships are calculated using big data foreign exchange investment VaR risk models
The maximum loss value of phase difference foreign exchange is cut, and suggests every foreign exchange transaction optimal delivery time, result of calculation is stored in data
In the mysql of storehouse, and showed in a manner of graph visualization.
The beneficial effects of the invention are as follows：By establishing big data foreign exchange investment VaR risk models, the different delivery times are assessed
Greatest risk penalty values of the different foreign exchanges under the fluctuation of market normal dynamic, carry out visual reference, enterprise can be in advance for enterprise
Calculate, reduce the market risk；It is determined that necessary capital and offer supervision foundation.
Brief description of the drawings
Fig. 1 is the method flow diagram using big data quantitative analysis foreign exchange investment risk of the present invention.
Embodiment
The present invention is directed to propose a kind of system and method using big data quantitative analysis foreign exchange investment risk, to assess not
With greatest risk penalty values of the delivery time difference foreign exchange under the fluctuation of market normal dynamic, visual reference is carried out for enterprise.
In specific implementation, the system in the present invention includes：Data acquisition module, data memory module, data processing mould
Block, big data foreign exchange investment VaR risk models algoritic module, model result output display module；
The data acquisition module, it is that bottom foreign currency operation system data, third party website data and part is manual
Data, database purchase is introduced by different modes；
The data memory module is to the initial data of collection and carried out the data of data processing and is associated and deposits
Storage, and support increment to store；
The data processing module, it is that the data demand proposed to foreign exchange business datum according to model carries out ETL processing；
The big data foreign exchange investment VaR risk model algoritic modules, for carrying out model foundation according to reduced data,
By continuing to optimize and machine learning, final output big data foreign exchange investment VaR risk models measure to quantifying riskization；
The model result exports display module, for by the measurement result of foreign exchange investment quantifying risk model with figure
Visual means are shown.
The method using big data quantitative analysis foreign exchange investment risk realized based on said system, the present invention, such as Fig. 1 institutes
Show, it includes implemented below step：
1 by bottom foreign currency operation system data, third party website data and part manual data, passes through different modes
Introduce database purchase；
In this step, asset management plateform system, SAP system, national exchange office enter mysql data by script interface
In storehouse；Third party's data excel is read and write in mysql by java program forms；Foreign exchange market information is grabbed by customization
The mode of page info reptile is taken to obtain in write-in mysql.
2. the data demand that foreign currency operation data propose according to model carries out ETL processing；
In this step, data demand is proposed by model group, carries out ETL work for the data in database mysql, as a result
Data continue to be stored in database to be used for model algorithm, specifically includes the place of data normalization processing and shortage of data value
Reason；Shortage of data value processing method has：1) predictive variable of missing is directly deleted；(2) using different methods to predictive variable
Missing values carry out interpolation, interpolating method has：Mean value interpolation, multiple interpolation, Random Interpolation, k nearest neighbor interpolation, linear interpolation etc..
3. the initial data of pair collection and carried out the data of data processing and be associated storage；
In this step, structuring and unstructured data (audio, image, e-mail, electrical form, txt texts to collection
Sheet, document, report) stored, and storage is associated to the data after data processing；
4. model foundation is carried out according to reduced data, by continuing to optimize and machine learning, outside final output big data
The investment VaR risk models that converge measure to quantifying riskization；
41st, index system is established：According to the factor for influenceing big data foreign exchange investment VaR risk models, service index system
The feature selecting flow of foundation carries out the screening and combination of index feature, and output is applied to big data foreign exchange investment VaR risk moulds
The index feature system of type；Index feature screening technique：(1) statistics such as correlation [threshold filtering]；(2) information gain, information
Ratio of profit increase, Gini coefficient etc. [threshold filtering]；(3) forward, backward and progressively back-and-forth method, such as AIC/BIC criterions [minimum value]；
(4) model selects, such as random forest, LASSO [model output].
42nd, model learning：Model obtained from database mysql user profile, accounts information, sequence information, procurement information,
Foreign exchange currency type, foreign exchange rate, systematic parameter, carry out big data foreign exchange investment VaR risk models and calculate, and by result input data
In the mysql of storehouse；
43rd, model optimization：By being continued to optimize to index, the continuous training to model, final big data foreign exchange is exported
Invest VaR risk models.The maximum loss value of different delivery time difference foreign exchanges is calculated, and suggests that every foreign exchange transaction is most preferably completed a business transaction
Time.
5. the measurement result of foreign exchange investment quantifying risk model is shown in a manner of graph visualization.
Claims (8)
1. utilize the system of big data quantitative analysis foreign exchange investment risk, it is characterised in that including：Data acquisition module, data
Memory module, data processing module, big data foreign exchange investment VaR risk models algoritic module, model result output display module；
The data acquisition module, be by bottom foreign currency operation system data, third party website data and part manual data,
Database purchase is introduced by different modes；
The data memory module is to the initial data of collection and carried out the data of data processing and is associated storage,
And increment is supported to store；
The data processing module, it is that the data demand proposed to foreign exchange business datum according to model carries out ETL processing；
The big data foreign exchange investment VaR risk model algoritic modules, for carrying out model foundation according to reduced data, pass through
Continue to optimize and machine learning, final output big data foreign exchange investment VaR risk models measure to quantifying riskization；
The model result exports display module, for the measurement result of foreign exchange investment quantifying risk model is visual with figure
Change mode is shown.
2. the system of big data quantitative analysis foreign exchange investment risk is utilized as claimed in claim 1, it is characterised in that described outer
The measurement result of remittance investment risk quantitative model includes：The maximum loss value of different delivery time difference foreign exchanges is calculated, and is suggested
Every foreign exchange transaction optimal delivery time.
3. utilize the method for big data quantitative analysis foreign exchange investment risk, it is characterised in that comprise the following steps：
A. by bottom foreign currency operation system data, third party website data and part manual data, drawn by different modes
Enter database purchase；
B. the data demand that foreign currency operation data propose according to model carries out ETL processing；
C. the initial data to collection and the data of data processing were carried out it is associated storage；
D. model foundation is carried out according to reduced data, by continuing to optimize and machine learning, final output big data foreign exchange is thrown
VaR risk models are provided to measure to quantifying riskization；
E. the measurement result of foreign exchange investment quantifying risk model is shown in a manner of graph visualization.
4. the method for big data quantitative analysis foreign exchange investment risk is utilized as claimed in claim 3, it is characterised in that step a
Specifically include：
Prepare data-interface access system internal user information, accounts information, sequence information, procurement information and outside national foreign exchange
Office data, then it is stored in database mysql；
Reptile obtains foreign exchange market industry and financial policy dynamic data, and extra festivals or holidays data deposit database mysql；
Third party's data excel is imported in database mysql by java program forms.
5. the method for big data quantitative analysis foreign exchange investment risk is utilized as claimed in claim 3, it is characterised in that step b
In, the ETL processing includes：Data needed for model are pre-processed to be used suitable for model, specifically include data standard
Change processing and the processing of shortage of data value.
6. the method for big data quantitative analysis foreign exchange investment risk is utilized as claimed in claim 3, it is characterised in that step c
In, the initial data of the collection includes structuring and non-structured data.
7. the method for big data quantitative analysis foreign exchange investment risk is utilized as claimed in claim 3, it is characterised in that step d
Specifically include：
D1, index system are established：According to the factor for influenceing big data foreign exchange investment VaR risk models, service index Establishing
Feature selecting flow carry out the screening and combination of index feature, output is suitable for big data foreign exchange investment VaR risk models
Index feature system；
D2, model learning：Model obtains user profile, accounts information, sequence information, procurement information, foreign exchange from database mysql
Currency type, foreign exchange rate, systematic parameter, carry out big data foreign exchange investment VaR risk models and calculate, and by result input database
In mysql；
D3, model optimization：By being continued to optimize to index, the continuous training to model, final big data foreign exchange investment is exported
VaR risk models.
8. the method for big data quantitative analysis foreign exchange investment risk is utilized as claimed in claim 3, it is characterised in that step e
In, in addition to：The maximum loss value of different delivery time difference foreign exchanges is calculated using big data foreign exchange investment VaR risk models, and
It is recommended that every foreign exchange transaction optimal delivery time, result of calculation is stored in database mysql, and in a manner of graph visualization
Show.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710875808.8A CN107609784A (en) | 2017-09-25 | 2017-09-25 | Utilize the system and method for big data quantitative analysis foreign exchange investment risk |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710875808.8A CN107609784A (en) | 2017-09-25 | 2017-09-25 | Utilize the system and method for big data quantitative analysis foreign exchange investment risk |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107609784A true CN107609784A (en) | 2018-01-19 |
Family
ID=61058288
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710875808.8A Pending CN107609784A (en) | 2017-09-25 | 2017-09-25 | Utilize the system and method for big data quantitative analysis foreign exchange investment risk |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107609784A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108038635A (en) * | 2018-02-01 | 2018-05-15 | 深圳云图智联技术有限公司 | The modeling of infrastructure assets investment repayment and analysis method and system |
| CN108062639A (en) * | 2018-02-23 | 2018-05-22 | 大连火眼征信管理有限公司 | A kind of Risk Propagation Model and the algorithm suitable for the model |
| CN109214925A (en) * | 2018-08-16 | 2019-01-15 | 深圳前海乘方互联网金融服务有限公司 | A kind of investment value assessment system |
| CN110400207A (en) * | 2019-07-31 | 2019-11-01 | 华北电力大学（保定） | A kind of financial On-line monitor system and method based on big data |
| CN112258095A (en) * | 2020-12-22 | 2021-01-22 | 中国平安财产保险股份有限公司 | Standard normal distribution based scoring method, device, equipment and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106127576A (en) * | 2016-07-01 | 2016-11-16 | 武汉泰迪智慧科技有限公司 | A kind of bank risk based on user behavior assessment system |
| CN106503853A (en) * | 2016-11-02 | 2017-03-15 | 华南师范大学 | A kind of foreign exchange transaction forecast model based on multiple scale convolutional neural networks |
| CN107103534A (en) * | 2017-01-25 | 2017-08-29 | 世纪禾光科技发展（北京）有限公司 | A kind of Foreign trade electronic commerce platform home currency that solves receives the method and system that single guaranteed value of foreign exchange transaction capital is completed a business transaction |
-
2017
- 2017-09-25 CN CN201710875808.8A patent/CN107609784A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106127576A (en) * | 2016-07-01 | 2016-11-16 | 武汉泰迪智慧科技有限公司 | A kind of bank risk based on user behavior assessment system |
| CN106503853A (en) * | 2016-11-02 | 2017-03-15 | 华南师范大学 | A kind of foreign exchange transaction forecast model based on multiple scale convolutional neural networks |
| CN107103534A (en) * | 2017-01-25 | 2017-08-29 | 世纪禾光科技发展（北京）有限公司 | A kind of Foreign trade electronic commerce platform home currency that solves receives the method and system that single guaranteed value of foreign exchange transaction capital is completed a business transaction |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108038635A (en) * | 2018-02-01 | 2018-05-15 | 深圳云图智联技术有限公司 | The modeling of infrastructure assets investment repayment and analysis method and system |
| CN108038635B (en) * | 2018-02-01 | 2022-04-19 | 深圳云图智联技术有限公司 | Modeling and analyzing method and system for return on investment of infrastructure assets |
| CN108062639A (en) * | 2018-02-23 | 2018-05-22 | 大连火眼征信管理有限公司 | A kind of Risk Propagation Model and the algorithm suitable for the model |
| CN109214925A (en) * | 2018-08-16 | 2019-01-15 | 深圳前海乘方互联网金融服务有限公司 | A kind of investment value assessment system |
| CN110400207A (en) * | 2019-07-31 | 2019-11-01 | 华北电力大学（保定） | A kind of financial On-line monitor system and method based on big data |
| CN112258095A (en) * | 2020-12-22 | 2021-01-22 | 中国平安财产保险股份有限公司 | Standard normal distribution based scoring method, device, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107609784A (en) | Utilize the system and method for big data quantitative analysis foreign exchange investment risk | |
| Cho et al. | The impact of financial technology on China’s banking industry: An application of the metafrontier cost Malmquist productivity index | |
| CN109509040A (en) | Predict modeling method, marketing method and the device of fund potential customers | |
| CN112598500A (en) | Credit processing method and system for non-limit client | |
| CN101071477A (en) | Financial analysis system and method based on expert system and nonlinear technology | |
| CN109740792A (en) | Data predication method, system, terminal and computer storage medium | |
| CN107103423A (en) | One kind integrates budgeted financial system | |
| CN102117469A (en) | System and method for estimating credit risks | |
| CN101739646A (en) | Systematic risk control method, system and computer program product thereof | |
| KR102620000B1 (en) | Method and device for providing investment valuation service based on financial indicator analysis | |
| CN116503174A (en) | Financial data prediction system based on big data | |
| Chu | Inflation, innovation, and growth: A survey | |
| AL-SLEHAT et al. | The relationship between non-interest revenue and sustainable growth rate: A case study of commercial banks in Jordan | |
| Zhang | Flow of funds analysis: Innovation and development | |
| Sasmita et al. | Analysis of Bank Soundness Levels: Review Report of PT. Bank Bukopin in 2016-2020 Period | |
| CN113421014A (en) | Target enterprise determination method, device, equipment and storage medium | |
| Rota | Credit and growth: reconsidering Italian industrial policy during the Golden Age | |
| CN107358307A (en) | A kind of power grid enterprises investment budgey method under multi-constraint condition based on electricity sales amount prediction | |
| CN116843446A (en) | Method for evaluating and attributing infraction risk of small amount credit based on interpretable machine learning | |
| Gamsakhurdia et al. | Why Cash Optimization is Critical in Georgian Companies | |
| Devereux et al. | The extensive margin, sectoral shares, and international business cycles | |
| CN101122991A (en) | The fourth generation finance core business system systematic structure and intelligent financial accounting method | |
| De Blas et al. | FDI in the Banking Sector | |
| CN111626572A (en) | Digital intelligent wind control method and related product | |
| Nkwo | The Impact of Effective Asset Management on Financial Performance of Nigerian Consumer Goods Corporations |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180119 |