CN109191283A - Method for prewarning risk and system - Google Patents
Method for prewarning risk and system Download PDFInfo
- Publication number
- CN109191283A CN109191283A CN201811003266.6A CN201811003266A CN109191283A CN 109191283 A CN109191283 A CN 109191283A CN 201811003266 A CN201811003266 A CN 201811003266A CN 109191283 A CN109191283 A CN 109191283A
- Authority
- CN
- China
- Prior art keywords
- index
- warning
- derivative
- value
- threshold value
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/03—Credit; Loans; Processing thereof
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
Abstract
The present invention relates to a kind of method for prewarning risk and systems, and the method comprising the steps of: constructing the derivative index of each process monitoring class index；Collect the data value of the derivative index of each of each process monitoring class index of monitored target；Set the threshold value of warning of each derivative index；The data value of each derivative index is compared with corresponding threshold value of warning, whether Risk-warning signal is issued according to the quantitative determination of the derivative index beyond threshold value of warning.By the method for the invention and system, can effectively capture the early warning signal of non-strong risk indicator, and it is perspective be higher than strong risk indicator, the pre-warning signal that such index issues can greatly improve whole recall rate, reduce overall misjudgment rate.
Description
Technical field
The present invention relates to technical field of data processing, in particular to a kind of method for prewarning risk and system.
Background technique
Perfect and fine post-loan management system can be caught by thoroughgoing and painstaking follow-up investigations and specialized analysis
The potential default risk of client is caught, issues Risk-warning signal after borrowing in time, so that bank can take timely measure neutralizing risk,
Utmostly retrieve a loss.
In entire post-loan management system, the generation from risk indicator to pre-warning signal is the part of core the most.At present
Loan after pre-warning signal generate based on qualitative method, it is more by post-loan management personnel's business experience, pass through setting strong signal
Rule periodically carries out detection risk, judges whether occurrence risk.High wind danger forewarning index is usually directly to be referred to the data of index
As soon as it is compared with fixed threshold value, once it is more than that threshold value triggering generates pre-warning signal.Have one using strong signal rule early warning
Fixed pre-alerting ability, but actually non-strong risk indicator also has certain early warning meaning, can monitor risk indicator certainly
The deterioration of body, however but lack the method for prewarning risk of non-direction of strong wind index at present.
Summary of the invention
It is an object of the invention to improve drawbacks described above, a kind of new method for prewarning risk and system are provided.
In order to achieve the above-mentioned object of the invention, the embodiment of the invention provides following technical schemes:
On the one hand, a kind of method for prewarning risk is provided in the embodiment of the present invention, comprising the following steps:
Construct the derivative index of one or more of each process monitoring class index；
The data value of the derivative index of each of each process monitoring class index of real-time collecting monitored target；
Set the threshold value of warning of each derivative index；
The data value of each derivative index is compared with corresponding threshold value of warning in real time, according to spreading out beyond threshold value of warning
Whether the quantitative determination of raw index issues Risk-warning signal.
Make in the scheme advanced optimized, the above method further comprises the steps of: the data value based on each derivative index, draws
The probability distribution graph of each derivative index；The threshold value of warning of each derivative index of setting, is specifically referred to based on each derivative
Target probability distribution graph sets the threshold value of warning of each derivative index.Actual data value generating probability distribution based on derivative index
Figure sets threshold value of warning further according to distribution map, realizes that threshold value of warning also follows index value dynamic change, can be further improved wind
The reliability of dangerous early warning.
On the other hand, a kind of Warning System is provided simultaneously in the embodiment of the present invention, comprised the following modules:
Derivative index constructing module, the derivative index of one or more for constructing each process monitoring class index；
Data collection module, the derivative index of each of each process monitoring class index for real-time collecting monitored target
Data value；
Threshold setting module, for setting the threshold value of warning of each derivative index；
Risk-warning module, for the data value of each derivative index to be compared with corresponding threshold value of warning in real time, root
Risk-warning signal whether is issued according to the quantitative determination of the derivative index beyond threshold value of warning.
In another aspect, the embodiment of the present invention provides a kind of computer-readable storage including computer-readable instruction simultaneously
Medium, the computer-readable instruction make processor execute the operation in method described in the embodiment of the present invention when executed.
In another aspect, the embodiment of the present invention provides a kind of electronic equipment simultaneously, comprising: memory stores program instruction；
Processor is connected with the memory, executes the program instruction in memory, realizes in method described in the embodiment of the present invention
The step of.
Compared with prior art, the present invention by can only the time series data of longitudinal comparison be converted into the list that can continuously monitor
Point index, and single-point index is reacted into its degree of risk in the form being distributed, the vertical analysis and transverse direction of index can be achieved at the same time
Analysis, and index will change over time, achieve the purpose that dynamic monitoring.Vertical analysis is that industry is general with horizontal analysis
Concept, point data is compared when vertical analysis refers to that the data of various time points and one are basic, at any time with data
Between the trend that changes, including trend analysis and rate of change analysis, using rate of change analysis in this programme, under normal circumstances,
It is not in big difference that the data target of each time point of enterprise changes each other, if differing greatly should arouse attention；Laterally
Analysis refers to an enterprise compared with other enterprises are on same time point, and many enterprises can be placed on to same standard and compared
Compared with.
By analysis, the method for the present invention can effectively capture the early warning signal of non-strong risk indicator, can nearly refer to for high wind
Mark provides effectively supplement, and perspective higher than strong risk indicator, can be when strong risk indicator not yet prompts pre-warning signal, effectively
The deterioration feature for capturing risk client, achievees the purpose that timely early warning, reduces loss for bank.Specifically, the application present invention
Afterwards, the pre-warning signal that such index issues can greatly improve whole recall rate.Recall rate is the general statistical in credit risk
Concept, what which calculated is the ratio that all " samples being correctly retrieved " account for all " samples that should be retrieved ", is measured
The effect of classifier, the ratio is bigger, illustrate overdue client that early warning goes out account for all overdue clients accounting it is bigger, model
Effect is better,Reduce overall misjudgment rate, False Rate is credit
General statistical concept in risk, what which calculated is overdue client's number of miss and the ratio of non-early warning client sum
Example,The accuracy of classifier is measured, the ratio is smaller, and explanation will be hospitable
Family misjudgement is that the accounting of overdue client is fewer, and the effect of model is better.Under normal conditions, recall rate can make simultaneously with False Rate
It is " looking for entirely " with, recall rate height, False Rate is low i.e. " look for ".And it can reach and averagely give warning in advance more than 180 days, and hit
The average aging of client is in M2+, that is, the overdue number of days of the actual average of the overdue client hit was at 60 days or more.
Detailed description of the invention
In order to illustrate the technical solution of the embodiments of the present invention more clearly, below will be to needed in the embodiment attached
Figure is briefly described, it should be understood that the following drawings illustrates only certain embodiments of the present invention, therefore is not construed as pair
The restriction of range for those of ordinary skill in the art without creative efforts, can also be according to this
A little attached drawings obtain other relevant attached drawings.
Fig. 1 is a kind of flow chart of method for prewarning risk described in present pre-ferred embodiments.
Fig. 2 (a)-(d) is respectively the probability distribution graph of different derivative indexs.
Fig. 3 (a)-(c) is respectively that the early warning line of the index of different risk classifications divides schematic diagram.
Fig. 4 is a kind of functional block diagram of the Warning System provided in the present embodiment.
Fig. 5 is the structural block diagram of a kind of electronic equipment provided in the present embodiment.
Specific embodiment
Below in conjunction with attached drawing in the embodiment of the present invention, technical solution in the embodiment of the present invention carries out clear, complete
Ground description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.Usually exist
The component of the embodiment of the present invention described and illustrated in attached drawing can be arranged and be designed with a variety of different configurations herein.Cause
This, is not intended to limit claimed invention to the detailed description of the embodiment of the present invention provided in the accompanying drawings below
Range, but it is merely representative of selected embodiment of the invention.Based on the embodiment of the present invention, those skilled in the art are not doing
Every other embodiment obtained under the premise of creative work out, shall fall within the protection scope of the present invention.
High wind danger index definition: strong risk indicator is known as status monitoring class index (A class), it is of interest that index is when current
The variation that point occurs can determine whether early warning and warning level by the variation of current time point state.Strong risk indicator covers
Fundamental breach event, great illegal, industrial and commercial judicial abnormal state and other significant adverse factors, easily lead to awarding credit assets
It is sorted into the signal of bad class, strong risk indicator for example has external warranty stroke count, Enterprise Application income, enterprise income tax
Tax liability etc..Such index can prompt corresponding warning level by setting threshold value of warning, and high wind danger forewarning index is mesh
Preceding common alarm mode is only drawn concept, is not illustrated in the present embodiment this.
Non- high wind danger index definition: non-strong risk indicator is known as process monitoring class index (B class), such index is compared to strong
Risk indicator, degree of risk slightly subtract, can not by the state at single time point change come clearly determine its whether early warning or early warning
Rank, therefore after requirement monitors the variation characteristic in such index certain period of time, comprehensive judgement whether touch early warning line and
It issues warning signal.
Weak risk indicator definition: weak risk indicator is known as indicating risk class index (C class), such index is single index
Variation cannot obviously represent the increase of customer default risk, need to form index pond by multiple indexs, by setting index weights
To indicate that customer default risk increases.Such index can be used for establishing customer action Rating Model, pre- in the form that client scores
Survey high risk client group.)
Only as an example, the non-strong risk indicator in part and weak risk indicator are listed in table 1 below.It should be noted that
It is that in table 1, " strong index levels " are classified as the index of B, as " non-strong risk indicator " described in the present invention, and the finger for being C
Mark is weak risk indicator, and method for early warning is not belonging to scope of the invention, in addition, also having A class index, only when practical application
It does not list in the table, such index, that is, previously mentioned " strong risk indicator ", method for early warning is also not belonging to model of the invention
Farmland.Strong risk indicator is derivative without carrying out index, because strong risk indicator, which refers to represent once occurring, very big default risk
Index, do not need continue observation the one section time variation tendency after determined again.
Table 1
Referring to Fig. 1, the present embodiment has illustratively provided a kind of method for prewarning risk, method includes the following steps:
Step 1, the derivative index of each process monitoring class index is constructed.
The purpose of derivative index construction is the index of consecutive variations being evolved into single-point index, respectively by spreading out to multiple
The monitoring of raw index, completes dynamic monitoring.Herein, derivative index construction includes: to calculate " spreading out for each process monitoring class index
Raw index ", that is, construct 1 current criteria value of each process monitoring class index and the index value change rate of several periods,
As an example, for example including T time point index value, a monthly variety ratio r1, three monthly variety ratio r3 and six monthly variety ratio r6, meter
It is as follows to calculate rule:
In above-mentioned calculating formula, T indicate T time point index value, T-1 indicate T-1 time point (i.e. T time point it is one month corresponding before
Time point) index value, T-3 indicate T-3 time point (i.e. T time point it is three months corresponding before time point) index value, T-6 indicate T-
6 time points (i.e. T time point it is six months corresponding before time point) index value.
By taking " Enterprise Application income " as an example:
| T | T-1 | T-3 | T-6 | r1 | r3 | r6 |
| 95196 | 46295 | 74520 | 78460 | 0.5136 | -0.2172 | -0.1758 |
Step 2, the risk classifications of Classification Index.
Since each index may be in different relationships from risk, it is therefore desirable to treat with a certain discrimination.To process monitoring class index
Classified (the derivative index that process monitoring class index is derived with it is classified using identical risk classifications), according to index
Following classification is divided into the relationship of risk:
(1) positive index, that is, be the bigger the better, be to be negatively correlated relationship with risk, is worth bigger, and it is smaller to represent risk, in threshold
Value takes the left margin part of probability distribution graph when dividing；Such as: " Enterprise Application income ", which calculates from taxation declaration table
On the enterprise marketing Revenue put on record, judge from economic significance, enterprise marketing income is higher, illustrates that the enterprise management condition is got over
Good, loan repayment capacity is better, then default risk is lower, therefore is positive index.
(2) negative sense index, i.e., it is the smaller the better, it is positive correlation with risk, is worth smaller, it is smaller to represent risk, in threshold
Value takes the right margin part of probability distribution graph when dividing；Such as: " external warranty stroke count (enterprise) ", which calculates enterprise
Enterprise's external guaranty stroke count that industry is shown from reference report, judges, enterprise's external guaranty stroke count is more, says from economic significance
The contingent liability of the bright enterprise is more, i.e., loan repayment capacity may be poorer, then default risk is higher, therefore is negative sense index.
(3) osculant index is the middle relationship that becomes with risk, be worth to walk toward two boundaries, and it is bigger to represent risk, is divided in threshold value
When take left and right two boundary part of probability distribution graph.Such as: " once affiliated party's quantity ", the index were calculating enterprise once
The quantity of affiliated party, judges from economic significance, and affiliated party's quantity of enterprise is very few, illustrates that company size is small, and business is carried out not
Power, loan repayment capacity is poor, then default risk is big, meanwhile, if affiliated party's quantity of enterprise is excessive, illustrate the wind that enterprise may face
A possibility that danger point is more, generates material risk event by affiliated party and causes the chain reaction of ontology enterprise are big, influence loan repayment capacity,
Then default risk is big.
Step 3, the data value of the derivative index of each of each monitoring index of monitored target is obtained in real time.
Only by taking " Enterprise Application income " index as an example, in existing customer group, the derivative of " Enterprise Application income " index refers to
Target data value is as shown in table 1 below:
Step 4, the probability distribution graph of each derivative index is drawn.
Before drawing distribution map, needs whether judge index value (numerical value of index) exceptional value occurs, judge exceptional value
Mode is as follows:
(1) if positive index:
(2) if negative sense index:
It is normal condition when only index value is " calculated value ", remaining situation is all exceptional value, and exceptional value in upper table
In, the value according to T value and r1, r3, r6 is positive and negative, and can be assigned a value of null value or special valuation (such as 999999 or 666666)
It hits the target after construction, in loan customer group, each derivative index of each index can draw a secondary probability
Distribution map, probability distribution graph are established based on normal index value, and distribution function is exactly that variable is less than or equal to some particular value a
Probability, if X to be regarded as to the coordinate of the random point on number axis, functional value of the distribution function F (a) at a means that X is fallen
Section (- ∞, a) on probability, i.e. F (a)=P { X≤a } represents risky under extreme probability.
The element of probability distribution graph mainly includes the following contents:
(1) data point: each of probability distribution graph data point represents change corresponding in every section of history cycle index
Rate.The number of data point depends on the number of the data sample amount (i.e. client measures) taken when Primary Stage Data is collected.
(2) reference axis: with the numerical value (index value of value and current point including change rate, i.e. T value) of the derivative index of finger
Value is axis of abscissas (X-axis), using the value of data volume as axis of ordinates (Y-axis).
Only with listed data instance in above-mentioned example table 1, the probability distribution graph of T, r1, r3, r6 of drafting are respectively such as
Shown in Fig. 2 (a)-(d).
The purpose for carrying out Risk-warning is not to see whether some client has refund risk, so generally can be only for this
Single client is monitored.However in this method, based on customers, number of the index value in some range in statistics customers
On the one hand amount had both realized monitoring while to all enterprises, on the other hand importantly, using customers as reference standard,
Determine whether the index value of single client belongs to "abnormal" inside current this crowd of clients, improves the accuracy of Risk-warning.Than
It is negative from the point of view of numerical value not counting big if the index value change rate of some client is -0.1, but if entire customers currently
Change rate is all -0.0001 or so, and-the 0.1 of the client just belongs to "abnormal", it is necessary to early warning.
Step 5, the risk classifications based on derivative index, set threshold value of warning.
Based on the historical data of all clients, above-mentioned derivative index is constructed, and draws out corresponding probability distribution graph respectively,
On the basis of probability distribution graph, early warning line (threshold value of warning) can be determined by following two method:
Method one: selected fixed quantile.Quantile is also known as quantile, refers to the probability distribution of a stochastic variable
Range is divided into the numerical point of several equal portions, and there are commonly median (i.e. two quantiles), quartile, percentiles etc..Quartile
What number referred to is exactly a point in continuous distribution function, this puts corresponding Probability p.If probability 0 < p < 1, stochastic variable X or it
The quantile Za of probability distribution refers to and meets condition p (X≤Za)=α real number.The case where no any business experience is oriented to
Under, initial 5% quantile can be selected as early warning line.Because saying from the statistical significance, 5% belongs to the general " tail of industry
Portion " definition.Specifically, then early warning line is corresponding index value at 5% quantile if positive index；If negative sense index,
Then early warning line is corresponding index value at 95% quantile；If osculant index, then early warning line is respectively upper and lower 5% quantile
Corresponding index value, as shown in Fig. 3 (a)-(c).
Method two: genetic algorithm global optimizing.In the enough situations of data volume, it can also be carried out by genetic algorithm complete
Office's search, finds early warning blacklist client's hit rate highest (early warning blacklist of derivative index which quantile in distribution map
Client's hit rate highest refers to that overdue client's number of hit is most), and pre-warning signal quantity it is relatively reasonable (can in conjunction with business pass through
It tests and is judged, usual pre-warning signal quantity is greater than the 25% of total customer quantity, then is determined as unreasonable.Such as: one shares 400
Name client, for an index, if there is 100 clients have issued pre-warning signal, it is clear that just unreasonable, early warning letter
It is number excessive, the workload of customer manager after borrowing can be significantly greatly increased, be unfavorable for the timely processing of pre-warning signal), then with the quantile
For threshold value of warning.
The setting rule of threshold line is as follows:
(1) using the highest quantile of index blacklist hit rate as the initial setting up of threshold line；
(2) threshold line selects newest sample data according to the accumulation of data volume, runs genetic algorithm again, optionally carries out
Periodically (such as monthly, quarterly) update.
The step of mode based on genetic algorithm global optimizing sets threshold value of warning include:
(1) input data；
(2) parameter is set, and parameter includes Population Size (population), crossover probability (cross_prob), makes a variation generally
Rate (mutation_prob), maximum number of iterations (generation), threshold value value range to be searched (threshold_min,
Threshold_max), threshold number (n, can value 3,6), searching times (k), sampling of data quality are than (times)；
(3) it is taken out according to the multiple of bad sample (bad sample refers to overdue client herein, and good sample refers to normal refund client)
Sample has been taken, and has used all bad samples, has constituted new data set；
(4) data set more than carries out the search of GA (genetic algorithm) optimal threshold, and it is corresponding to export minimum fitness function
All optimal solutions (there may be multiple optimal solutions), and average.Fitness function is the error rate for adding penalty term, by bad person
Misjudgement is that the punishment of good person is 2, and good person's misjudgement is that the punishment of bad person is 1.Penalty term only focuses on the ratio of the two, with specific value
Size is unrelated；
(5) it after repeating step (3)-(4) k time, averages to the optimal value of k search, which is optimal threshold value of warning,
And calculate quantile, that is, determine the optimal threshold value of warning in probability distribution graph in where.
Step 6, the data value of each derivative index is compared with corresponding threshold value of warning in real time, according to beyond early warning threshold
Whether the quantitative determination of the derivative index of value issues Risk-warning signal.
After selecting the threshold value of warning in each index distribution map, if index value is greater than (or being less than) threshold value of warning, that is, fall
Enter " abnormal section ", if " the derivative index " of certain monitoring index reaches certain amount and fall into " abnormal section ", which is just triggered
In early warning, such as 4 derivative indexs, if T value is fallen into " abnormal section ", the direct early warning of the index, if T value does not fall within " exception
Section ", but there are 2 or more to fall into " abnormal section " in remaining 3 change rates, then the index issues warning signal.
Such as " taxation declaration income " this index, 4 derivative indexs can be derived, are exactly T value, r1, r3, r6, such as
Fruit T value falls into abnormal section, then " taxation declaration income " this index issues warning signal, if T value does not fall within exception
Section then has 2 or more to fall into abnormal section inside r1, r3, r6, then " taxation declaration income " this index issues early warning letter
Number.
Referring to Fig. 4, being based on identical inventive concept, a kind of Warning System is additionally provided in the present embodiment, this is
It is not directed to place in system description, referring again to the related content of preceding method description.
Specifically, the Warning System comprises the following modules:
Derivative index constructing module, the derivative index of one or more for constructing each monitoring index；
Data collection module, the data of the derivative index of each of each monitoring index for real-time collecting monitored target
Value；
Threshold setting module sets the threshold value of warning of each derivative index for the risk classifications according to derivative index；
Risk-warning module, for the data value of each derivative index to be compared with corresponding threshold value of warning in real time, root
Risk-warning signal whether is issued according to the quantitative determination of the derivative index beyond threshold value of warning.
In preferable embodiment, above-mentioned Warning System further includes graphic plotting module, for being received based on data
The data value for collecting the derivative index of each of module acquisition, draws the probability distribution graph of each derivative index.At this point, mould is arranged in threshold value
Block is then the threshold value of warning that the probability distribution graph based on each derivative index sets each derivative index.For example, when being based on current
Between the probability distribution graph of each derivative index of " Enterprise Application income " index that obtains of point, set 5% quantile as in advance
Alert line.Threshold value of warning is configured based on the probability distribution graph obtained in real time, either with fixed value setting early warning line, or with
Genetic algorithm obtain optimal value as threshold value of warning, threshold value of warning be all according to derivative index variation and dynamic change,
Then more effective reasonable early warning value can be set out, realize more accurately Risk-warning, enhance the reliable of Warning System
Property.
As shown in figure 5, the present embodiment provides a kind of electronic equipment simultaneously, which may include 51 He of processor
Memory 52, wherein memory 52 is coupled to processor 51.It is worth noting that, the figure is exemplary, it can also be used
The structure is supplemented or substituted to the structure of his type.
As shown in figure 5, the electronic equipment can also include: input unit 53, display unit 54 and power supply 55.It is worth noting
, which is also not necessary to include all components shown in Fig. 5.In addition, electronic equipment can also include
The component being not shown in Fig. 5 can refer to the prior art.
Processor 51 is sometimes referred to as controller or operational controls, may include microprocessor or other processor devices and/
Or logic device, the processor 51 receive the operation of all parts of input and controlling electronic devices.
Wherein, memory 52 for example can be buffer, flash memory, hard disk driver, removable medium, volatile memory, it is non-easily
The property lost one of memory or other appropriate devices or a variety of, can store configuration information, the processor 51 of above-mentioned processor 51
The instruction of execution, record the information such as list data.Processor 51 can execute the program of the storage of memory 52, to realize information
Storage or processing etc..It in one embodiment, further include buffer storage in memory 52, i.e. buffer, with the intermediate letter of storage
Breath.
Input unit 53 is for example for providing the data value of each derivative index of acquisition to be performed to processor 51.Display
Unit 54 is used to show various in treatment process that the data value as a result, for example each derivative index, the display unit for example may be used
Think LCD display, but the present invention is not limited thereto.Power supply 55 is used to provide electric power for electronic equipment.
The embodiment of the present invention also provides a kind of computer-readable instruction, wherein when executing described instruction in the electronic device
When, described program makes electronic equipment execute the operating procedure that the method for the present invention is included.
The embodiment of the present invention also provides a kind of storage medium for being stored with computer-readable instruction, wherein the computer can
Reading instruction makes electronic equipment execute the operating procedure that the method for the present invention is included.
It should be understood that in various embodiments of the present invention, magnitude of the sequence numbers of the above procedures are not meant to execute suitable
Sequence it is successive, the execution of each process sequence should be determined by its function and internal logic, the implementation without coping with the embodiment of the present invention
Process constitutes any restriction.
Those of ordinary skill in the art may be aware that list described in conjunction with the examples disclosed in the embodiments of the present disclosure
Member and algorithm steps, can be realized with electronic hardware, computer software, or a combination of the two, in order to clearly demonstrate hardware
With the interchangeability of software, each exemplary composition and step are generally described according to function in the above description.This
A little functions are implemented in hardware or software actually, the specific application and design constraint depending on technical solution.Specially
Industry technical staff can use different methods to achieve the described function each specific application, but this realization is not
It is considered as beyond the scope of this invention.
In several embodiments provided herein, it should be understood that disclosed systems, devices and methods, it can be with
It realizes by another way.For example, the apparatus embodiments described above are merely exemplary, for example, the unit
It divides, only a kind of logical function partition, there may be another division manner in actual implementation, such as multiple units or components
It can be combined or can be integrated into another system, or some features can be ignored or not executed.
If the integrated unit is realized in the form of SFU software functional unit and sells or use as independent product
When, it can store in a computer readable storage medium.Based on this understanding, technical solution of the present invention is substantially
The all or part of the part that contributes to existing technology or the technical solution can be in the form of software products in other words
It embodies, which is stored in a storage medium, including some instructions are used so that a computer
Equipment (can be personal computer, server or the network equipment etc.) executes the complete of each embodiment the method for the present invention
Portion or part steps.And storage medium above-mentioned includes: USB flash disk, mobile hard disk, read-only memory (ROM, Read-Only
Memory), random access memory (RAM, RandomAccess Memory), magnetic or disk etc. are various can store journey
The medium of sequence code.
The above description is merely a specific embodiment, but scope of protection of the present invention is not limited thereto, any
Those familiar with the art in the technical scope disclosed by the present invention, can easily think of the change or the replacement, and should all contain
Lid is within protection scope of the present invention.Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (11)
1. a kind of method for prewarning risk, which comprises the following steps:
Construct the derivative index of one or more of each monitoring index；
The data value of the derivative index of each of each monitoring index of real-time collecting monitored target；
Set the threshold value of warning of each derivative index；
The data value of each derivative index is compared with corresponding threshold value of warning in real time, is referred to according to the derivative beyond threshold value of warning
Whether target quantitative determination issues Risk-warning signal.
2. the method according to claim 1, wherein the one or more of each monitoring index of construction is derivative
The step of index, comprising: be directed to each monitoring index, construct the index a current point in time index value and several
The index value change rate of period.
3. the method according to claim 1, wherein further comprise the steps of: the data value based on each derivative index,
Draw the probability distribution graph of each derivative index；The threshold value of warning of each derivative index of setting, is specifically based on each spreading out
The probability distribution graph of raw index sets the threshold value of warning of each derivative index.
4. according to the method described in claim 3, it is characterized in that, the step of the threshold value of warning of each derivative index of setting
Suddenly, comprising: one fixed value of setting is as threshold value of warning.
5. according to the method described in claim 3, it is characterized in that, the step of the threshold value of warning of each derivative index of setting
Suddenly, comprising: using the highest quantile of index blacklist hit rate as the initial value of threshold line, be based on genetic algorithm News Search
Optimal value, using the optimal value as threshold value of warning.
6. according to the method described in claim 2, it is characterized in that, the quantity of derivative index of the basis beyond threshold value of warning
The step of determining whether to issue Risk-warning signal, comprising:
If the data value of current point in time index value exceeds corresponding threshold value of warning, Risk-warning signal is issued；
If the data value of current point in time index value is without departing from corresponding threshold value of warning, but the index value of several periods changes
In the data value of rate, has at least two beyond corresponding threshold value of warning, then issue Risk-warning signal.
7. the method according to claim 1, wherein it is corresponding pre- to judge whether the data value of derivative index exceeds
The step of alert threshold value, comprising:
Judge that the risk classifications of the derivative index, the risk classifications include positive index, negative sense index and osculant index；
If the derivative index is that positive index determines if the data value of the derivative index is greater than corresponding threshold value of warning
Exceed corresponding threshold value of warning for the data value of the derivative index；
If the derivative index is that negative sense index determines if the data value of the derivative index is less than corresponding threshold value of warning
Exceed corresponding threshold value of warning for the data value of the derivative index；
If the derivative index be osculant index, if the data value of the derivative index be less than smaller threshold value of warning and be greater than compared with
Big threshold value is then determined as the data value of the derivative index beyond corresponding threshold value of warning.
8. a kind of Warning System, which is characterized in that comprise the following modules:
Derivative index constructing module, the derivative index of one or more for constructing each monitoring index；
Data collection module, the data value of the derivative index of each of each monitoring index for real-time collecting monitored target；
Threshold setting module, for setting the threshold value of warning of each derivative index；
Risk-warning module, in real time comparing the data value of each derivative index with corresponding threshold value of warning, according to super
Whether the quantitative determination of the derivative index of threshold value of warning issues Risk-warning signal out.
9. system according to claim 8, which is characterized in that further include graphic plotting module, for being based on each derivative
The data value of index draws the probability distribution graph of each derivative index.
10. a kind of computer readable storage medium including computer-readable instruction, which is characterized in that the computer-readable finger
Enable the operation for requiring processor perform claim in any the method for 1-7.
11. a kind of electronic equipment, which is characterized in that the equipment includes:
Memory stores program instruction；
Processor is connected with the memory, executes the program instruction in memory, realizes that claim 1-7 is any described
Step in method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811003266.6A CN109191283A (en) | 2018-08-30 | 2018-08-30 | Method for prewarning risk and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811003266.6A CN109191283A (en) | 2018-08-30 | 2018-08-30 | Method for prewarning risk and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109191283A true CN109191283A (en) | 2019-01-11 |
Family
ID=64917443
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811003266.6A Pending CN109191283A (en) | 2018-08-30 | 2018-08-30 | Method for prewarning risk and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109191283A (en) |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110057310A (en) * | 2019-05-21 | 2019-07-26 | 中国能源建设集团陕西省电力设计院有限公司 | A kind of artificial digging pile deformation monitoring alarm method and system |
| CN110223167A (en) * | 2019-06-19 | 2019-09-10 | 中国工商银行股份有限公司 | Production transaction cooperative monitoring method and production transaction association monitoring method |
| CN110320872A (en) * | 2019-06-28 | 2019-10-11 | 云南中烟工业有限责任公司 | A method of improving cigarette main stream flue gas index standardized production |
| CN110335136A (en) * | 2019-06-03 | 2019-10-15 | 中国民生银行股份有限公司 | Data processing method and device |
| CN110349002A (en) * | 2019-07-02 | 2019-10-18 | 北京淇瑀信息科技有限公司 | A kind of method, apparatus and electronic equipment of consumer finance whole process monitoring and early warning |
| CN110751371A (en) * | 2019-09-20 | 2020-02-04 | 苏宁云计算有限公司 | Commodity inventory risk early warning method and system based on statistical four-bit distance and computer readable storage medium |
| CN111192121A (en) * | 2019-12-17 | 2020-05-22 | 航天信息股份有限公司 | ANN-based automatic risk taxpayer early warning method and system |
| CN111461892A (en) * | 2020-03-31 | 2020-07-28 | 支付宝(杭州)信息技术有限公司 | Method and device for selecting derived variables of risk identification model |
| CN111861730A (en) * | 2020-07-31 | 2020-10-30 | 重庆富民银行股份有限公司 | Risk report early warning system and method |
| CN111967767A (en) * | 2020-08-18 | 2020-11-20 | 支付宝(杭州)信息技术有限公司 | Business risk identification method, device, equipment and medium |
| CN112488569A (en) * | 2020-12-11 | 2021-03-12 | 绿瘦健康产业集团有限公司 | Risk early warning method and device and storage medium |
| CN113065736A (en) * | 2021-01-15 | 2021-07-02 | 开鑫金服(南京)信息服务有限公司 | Risk monitoring method, system, equipment and storage medium based on supply chain system |
| WO2021164465A1 (en) * | 2020-02-20 | 2021-08-26 | 深圳壹账通智能科技有限公司 | Intelligent early warning method and system |
| CN113920381A (en) * | 2021-12-15 | 2022-01-11 | 深圳市明源云科技有限公司 | Repeated derivative index identification method, electronic device and readable storage medium |
| CN117853081A (en) * | 2024-02-06 | 2024-04-09 | 一智科技(成都)有限公司 | Worker's early warning system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102238465A (en) * | 2010-04-28 | 2011-11-09 | 北京神州泰岳软件股份有限公司 | Mobile data service end-to-end quality analysis method and system |
| CN102751725A (en) * | 2012-07-13 | 2012-10-24 | 中国电力科学研究院 | Overload risk state identifying method for power distribution network |
| CN104834984A (en) * | 2015-02-11 | 2015-08-12 | 国家电网公司 | Electric power transaction supervision risk early warning system based on unified and interconnected electric power market |
| CN108230143A (en) * | 2018-02-05 | 2018-06-29 | 深圳微众税银信息服务有限公司 | Data report generation method and system after a kind of loan |
| CN112380089A (en) * | 2020-11-10 | 2021-02-19 | 深圳供电局有限公司 | Data center monitoring and early warning method and system |
-
2018
- 2018-08-30 CN CN201811003266.6A patent/CN109191283A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102238465A (en) * | 2010-04-28 | 2011-11-09 | 北京神州泰岳软件股份有限公司 | Mobile data service end-to-end quality analysis method and system |
| CN102751725A (en) * | 2012-07-13 | 2012-10-24 | 中国电力科学研究院 | Overload risk state identifying method for power distribution network |
| CN104834984A (en) * | 2015-02-11 | 2015-08-12 | 国家电网公司 | Electric power transaction supervision risk early warning system based on unified and interconnected electric power market |
| CN108230143A (en) * | 2018-02-05 | 2018-06-29 | 深圳微众税银信息服务有限公司 | Data report generation method and system after a kind of loan |
| CN112380089A (en) * | 2020-11-10 | 2021-02-19 | 深圳供电局有限公司 | Data center monitoring and early warning method and system |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110057310A (en) * | 2019-05-21 | 2019-07-26 | 中国能源建设集团陕西省电力设计院有限公司 | A kind of artificial digging pile deformation monitoring alarm method and system |
| CN110335136A (en) * | 2019-06-03 | 2019-10-15 | 中国民生银行股份有限公司 | Data processing method and device |
| CN110223167A (en) * | 2019-06-19 | 2019-09-10 | 中国工商银行股份有限公司 | Production transaction cooperative monitoring method and production transaction association monitoring method |
| CN110320872B (en) * | 2019-06-28 | 2021-08-03 | 云南中烟工业有限责任公司 | Method for improving standard production of cigarette mainstream smoke indexes |
| CN110320872A (en) * | 2019-06-28 | 2019-10-11 | 云南中烟工业有限责任公司 | A method of improving cigarette main stream flue gas index standardized production |
| CN110349002A (en) * | 2019-07-02 | 2019-10-18 | 北京淇瑀信息科技有限公司 | A kind of method, apparatus and electronic equipment of consumer finance whole process monitoring and early warning |
| CN110349002B (en) * | 2019-07-02 | 2024-01-16 | 北京淇瑀信息科技有限公司 | Method and device for monitoring and early warning of whole flow of consumption finance and electronic equipment |
| CN110751371A (en) * | 2019-09-20 | 2020-02-04 | 苏宁云计算有限公司 | Commodity inventory risk early warning method and system based on statistical four-bit distance and computer readable storage medium |
| CN111192121A (en) * | 2019-12-17 | 2020-05-22 | 航天信息股份有限公司 | ANN-based automatic risk taxpayer early warning method and system |
| WO2021164465A1 (en) * | 2020-02-20 | 2021-08-26 | 深圳壹账通智能科技有限公司 | Intelligent early warning method and system |
| CN111461892A (en) * | 2020-03-31 | 2020-07-28 | 支付宝(杭州)信息技术有限公司 | Method and device for selecting derived variables of risk identification model |
| CN111861730A (en) * | 2020-07-31 | 2020-10-30 | 重庆富民银行股份有限公司 | Risk report early warning system and method |
| CN111967767A (en) * | 2020-08-18 | 2020-11-20 | 支付宝(杭州)信息技术有限公司 | Business risk identification method, device, equipment and medium |
| CN112488569A (en) * | 2020-12-11 | 2021-03-12 | 绿瘦健康产业集团有限公司 | Risk early warning method and device and storage medium |
| CN113065736A (en) * | 2021-01-15 | 2021-07-02 | 开鑫金服(南京)信息服务有限公司 | Risk monitoring method, system, equipment and storage medium based on supply chain system |
| CN113920381A (en) * | 2021-12-15 | 2022-01-11 | 深圳市明源云科技有限公司 | Repeated derivative index identification method, electronic device and readable storage medium |
| CN113920381B (en) * | 2021-12-15 | 2022-04-15 | 深圳市明源云科技有限公司 | Repeated derivative index identification method, electronic device and readable storage medium |
| CN117853081A (en) * | 2024-02-06 | 2024-04-09 | 一智科技(成都)有限公司 | Worker's early warning system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109191283A (en) | Method for prewarning risk and system | |
| Burkart et al. | Leading indicators of currency crises for emerging countries | |
| CN104881783A (en) | E-bank account fraudulent conduct and risk detecting method and system | |
| US10614073B2 (en) | System and method for using data incident based modeling and prediction | |
| CN111553563A (en) | Method and device for determining enterprise fraud risk | |
| CN112102076A (en) | Comprehensive risk early warning system of platform | |
| CN113177839A (en) | Credit risk assessment method, device, storage medium and equipment | |
| Wang et al. | Can gold hedge against oil price movements: Evidence from GARCH-EVT wavelet modeling | |
| CN111310803B (en) | Environment data processing method and device | |
| CN112488496A (en) | Financial index prediction method and device | |
| CN114742402A (en) | Information monitoring method, device, equipment and medium | |
| CN115063228A (en) | Anti-money laundering management platform and anti-money laundering method | |
| CN111222968A (en) | Enterprise tax risk management and control method and system | |
| CN111951105A (en) | Intelligent credit wind control system based on multidimensional big data analysis | |
| CN112037006A (en) | Credit risk identification method and device for small and micro enterprises | |
| CN113835947A (en) | Method and system for determining abnormality reason based on abnormality identification result | |
| Gerlach et al. | Dissection of Bitcoin's Multiscale Bubble History | |
| KR20130142353A (en) | System for monitoring client | |
| KR102336462B1 (en) | Apparatus and method of credit rating | |
| CN115689713A (en) | Abnormal risk data processing method and device, computer equipment and storage medium | |
| CN110675240B (en) | Monitoring method and system for risk radar early warning | |
| CN114254857A (en) | Power equipment inventory condition evaluation method and server | |
| CN114626940A (en) | Data analysis method and device and electronic equipment | |
| CN113435748A (en) | Dot state determination method and apparatus, electronic device and storage medium | |
| KR101954131B1 (en) | An unmanned payment system |
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 |
Application publication date: 20190111 |
|
| RJ01 | Rejection of invention patent application after publication |