CN102436622A - Method for evaluating network market operator credit status - Google Patents
Method for evaluating network market operator credit status Download PDFInfo
- Publication number
- CN102436622A CN102436622A CN2011104460532A CN201110446053A CN102436622A CN 102436622 A CN102436622 A CN 102436622A CN 2011104460532 A CN2011104460532 A CN 2011104460532A CN 201110446053 A CN201110446053 A CN 201110446053A CN 102436622 A CN102436622 A CN 102436622A
- Authority
- CN
- China
- Prior art keywords
- index
- twos
- credit
- judgment matrix
- layer
- 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
Images
Landscapes
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The invention discloses a method for evaluating the network market operator credit status. The method adopts a network market operator credit evaluation index system, so that the method inherits a classic evaluation model in the financial industry and combines with the self characteristic of an on-line trading market to reload, deform and extend, and the content and the amount of covered aspects and investigation dimensions are reasonable and efficient. Meanwhile, the invention also adopts an AHP (Analytic Hierarchy Process) method to determine an index system weight. After a judgment matrix is determined on the basis of an expert questionnaire, consistency checkout is carried out so as to reduce and lower the influence of an objective factor, and the reliability of the determined weight is guaranteed. In addition, the invention also introduces in the data of the outside authority of an on-line trading market, a bank and the state administration of industry and commerce, and the defect of poor evaluation authority and comprehensiveness in the prior art can be overcome.
Description
Technical field
The invention belongs to e-commerce field, particularly a kind of network market operator's credit assessment method.
Background technology
Ecommerce small enterprise is meant and utilizes Internet technology and channel, participates in the small enterprise of network trading.Since the nineties in 20th century, the develop rapidly of network economy and ecommerce makes social economy's pattern that deep change take place, and the internet begins to be penetrated into each corner of social production and life.Ecommerce is as novel business model, and network market operator is important participant, and credit then is basis and the support that ecommerce develops in a healthy way.How effective evaluation network market operator's credit becomes urgent and problem reality.
From traditional credit appraisal angle, mainly contain qualitative and quantitative two kinds of methods.Typical quilitative method is the 5C analytic approach, pays close attention to moral popularity, capital strength, loan repayment capacity, guarantee and five aspects of business environment of estimating object.Method for quantitatively evaluating is then mainly set about from the statistical study equal angles; Researchs such as woods C. Thomas are thought; The statistical method of credit appraisal is the techniques of discriminant analysis that Fei Sheer proposed in 1936 at first, and afterwards, Logistic returns became the most general statistical method of credit appraisal field use; Over nearest more than 20 years, the classification tree method enjoys favor.
In the electric commerce credit assessment field, most Application Research statistics and non-statistical method, exploration has been made in credit appraisal.L.Mui etc. understand on the basis at trust, prestige and reciprocal sociology, biology, have designed a kind of computation model, calculate network agent merchant's prestige mark; A.Jsang etc. have proposed a kind of credit scoring measure based on internet business; Wu Chong etc. propose to utilize the SVMs integrated approach based on fuzzy integral, and client's credit under the e-commerce environment is assessed; Peng Lifang etc. think in the e-commerce transaction, can obtain parties credit in the behavior of transactions history in the past through the accumulative total parties.On the whole, above-mentioned research lays particular emphasis on the evaluation of transaction itself more, and the authority of evaluation and comprehensive existence are not enough.
Analytical hierarchy process (Analytic Hierarchy Process) is called for short AHP; Be that the U.S. scholar T.L.Saaty that plans strategies for is taught in the initial stage seventies and proposes, AHP is a kind of easy, the flexible and practical multiple criteria decision making (MCDM) method of qualitative question being carried out quantitative test.Its feature is through being divided into the orderly level that connects each other the various factors in the challenge; Make it methodization; According to certain extension subjective judgement structure (mainly being to compare in twos) directly and is effectively combined expert opinion and analyst's objective judged result, the importance that each hierarchical elements is compared in twos is quantitatively described.Then, utilize mathematical method to calculate the weights of the relative importance order of each hierarchical elements of reflection, through the relative weighting of the total sorting calculation all elements between all levels line ordering of going forward side by side.
Its basic characteristics are: (1) systematicness.The AHP method is regarded object as system, according to decompose, relatively judge, the comprehensive mode of thinking makes a strategic decision; (2) practicality.Qualitative and quantitative is wanted to combine, and can handle the indeterminable problem of traditional optimization methods.(3) brief introduction property.Calculate easyly, the result is clear and definite, is convenient to the decision maker and is directly acquainted with and grasps.
The algorithm of analytical hierarchy process is following: the step analysis structural model is set up in (1).Analyse in depth practical problems, with related factors layering from top to bottom (target-criterion or index-scheme or object), lower floor is influenced by the upper strata, and each factor is relatively independent in every layer.(2) be configured to contrast matrix, construct the paired comparator matrix of each layer each factor of last layer with Paired Comparisons and 1-9 yardstick.(3) calculate weight vector and do consistency check, that is, to each contrast matrix, calculate maximum characteristic root and proper vector, do consistency check, if pass through, then proper vector is a weight vector.
Summary of the invention
The invention provides a kind of network market operator's credit assessment method, introduced the evaluation information of outside authoritative institution, thereby overcome the authoritative and comprehensive defect of insufficient of the evaluation that exists in the prior art enterprise.
A kind of network market operator's credit assessment method may further comprise the steps:
(1) confirms network market operator credit appraisal index system; Described network market operator's credit appraisal index system is made up of destination layer, rule layer and factor layer; Described destination layer is a network market operator credit appraisal index; Described rule layer comprises operator's quality index, management ability index, online transaction feedback information index, Internet business environmental baseline index, capital strength's index, continuity and public letter record index; Described factor layer is that each the subordinate's index by said rule layer constitutes; Wherein, Said operator's quality index is made up of nature person's credit card, legal person's credit rating, charitable activity and honor, educational background and professional standards, each index of political affiliation; Said management ability index by the operation life of abiding by the law continuously, credit evaluation result over the years, be engaged in Internet business headcount, Network employee wage is counted each index and is constituted; Said online transaction feedback information index is made up of transaction sequential statistics, dealing money sequential statistics, each index of transaction results character property evaluation; Said Internet business environmental baseline index is made up of online sign, network trading market establishment time, management place, each index of annual test situation over the years; Said capital strength index is made up of registered capital, asset-liability ratio, number of stock turnover, the accounts receivable number of turnover, each index of sales revenue rate of growth, and said continuity and public letter record index are made up of industrial and commercial credit rating, bank credit grading, black/gray list, criminal penalty information, tax credit rating, each index of customs's credit rating; It is thus clear that described network market operator's credit appraisal index system has contained 6 aspects, be provided with 27 investigation dimensions altogether;
(2) obtain data from network trading market, bank and State Administration for Industry & Commerce; And mate through unique sign of network market operator, obtain the data of each network market operator each index in described network market operator's credit appraisal index system; And the data of each index all are divided into 5 grades, like this, each data in described network market operator's credit appraisal index system just can be mapped in [0,5] interval.
(3) utilize analytical hierarchy process (AHP method) to confirm the weight of each index in the index system:
(3.1) construct judgment matrix in twos;
" 1～9 ratio scale method " table that adopts TLSatty to propose compares according to the relative importance of each index in the same level for last layer correlations index in twos, constructs judgment matrix in twos; Comprise:
Compare in twos according to the relative importance of each index in the rule layer for destination layer, the structure rule layer is judgment matrix in twos;
Compare in twos according to the relative importance of each index in the factor layer for index of correlation in the last rule layer, the structural factor layer is judgment matrix in twos;
Specific as follows:
The A level is the last layer of B level, as: A is a destination layer, and B is a rule layer; Perhaps A is a rule layer, and B is a factor layer.If be in each the index (B in the B level together
1, B
2..., B
n) with the A level in a certain index A
kRelevant, then B level judgment matrix can be expressed as:
Wherein, b
Ij＞0, b
Ii=1, b
Ij=1/b
Ji, i, j=1,2 ..., n; b
IjExpression is to a certain index A in the A level
k, B
iIndex is to B
jThe numeric representation of the relative importance of index;
(3.2) calculate the proper vector of said judgment matrix in twos, obtain the weight of each index in the index system;
For the judgment matrix in twos of n dimension, calculate the n th root of each row all elements product in the judgment matrix in twos, that is:
Wherein, b
IjElement for the capable j row of i in the judgment matrix in twos;
Do normalization to these root vectors of trying to achieve again and handle, obtain i characteristic vector W
i, computing formula is:
Thereby, obtain characteristic vector W=[w
1, w
2..., w
n]
T, be the weight of each index.
(3.3) calculate the coincident indicator C.I. of said judgment matrix in twos and Consistency Ratio C.R. at random, check the above-mentioned respectively consistance of judgment matrix in twos: when C.I.=0, judgment matrix has crash consistency in twos, is regarded as having satisfied consistance; When C.I. is not 0, judge that according to the value of C.R. when C.R.＜0.1, judgment matrix has satisfied consistance in twos, otherwise, just must readjust the element in the judgment matrix in twos, till judgment matrix has satisfied consistance in twos;
Wherein, described coincident indicator C.I. calculates through following formula:
λ max is the n rank eigenvalue of maximum of judgment matrix in twos,
Wherein, B is judgment matrix, W
iBe i proper vector, W is a proper vector.
The described C.R. of Consistency Ratio at random calculates through following formula:
R.I. be mean random coincident indicator in the analytical hierarchy process (AHP method).
(4) network market operator is carried out credit comprehensive evaluation:
According to the data of each index in the step (2) and the weight of each index in the step (3); Calculate the score of each index in described network market operator's credit appraisal index system; According between the scoring area that each score value dropped on; Its concrete implication is carried out semantic interpretation, and finally provide a credit evaluation report.
The present invention adopts network market operator credit appraisal index system; Both inherited the classical assessment models of financial industry; Combine the self-characteristic in network trading market to carry out heavy duty, distortion and extension again, contain aspect and the content of investigating dimension and reasonable quantity, efficient; Simultaneously, the present invention has also adopted the AHP method to confirm the index system weight, after confirming judgment matrix on the basis of expert's questionnaire, carries out consistency check again, thereby reduces the influence of subjective factor, guarantees the reliability of determined weight.
The consistance aspect is considered internally, adopts Cronbach ' s side reaction coefficient method commonly used at present that the reliability of the inventive method is tested.Reliability (Reliability) is a reliability, and it is meant the degree of consistency of adopting gained result when using the same method to the same target duplicate measurements.And according to the research of J.M.Cortina, the entry number of scale is very big to the side reaction coefficient influence, when the average related levels of clauses and subclauses is low, is like this especially.J.M.Cortina also thinks, if scale surpasses 14 clauses and subclauses, even then two orthogonal dimensions to comprise interrelated clauses and subclauses less, side reaction coefficient also can reach 0.70 even higher.In view of factor layer index in network market operator's credit appraisal index system of the inventive method reaches 27, the present invention is that the rule layer index is carried out check when adopting Cronbach ' s side reaction coefficient method.The result shows that total side reaction coefficient is 0.809, and the reliability of index system is better.
In addition, the validity of the inventive method is also tested.Validity is a validity, is meant that survey instrument or means can accurately measure the degree of the things of required measurement.Adopt factorial analysis that the construction validity of rule layer index score is tested.Before the factorial analysis, carried out that KMO estimates and Bartlett ' s sphericity test, the result shows that the KMO value is 0.855, and the factorial analysis condition is satisfied in Bartlett ' s sphericity test P＜0.001.Adopt main composition method, extract 1 common factor and analyze, the result shows that variance contribution ratio reaches 60.156%, and information extraction validity is better.Further, also carried out the cross check.According to the traditional bank rating result; 100 tame electronics commercial affairs small enterprise sample can be divided into medium and two groups of above, low middlings; Adopt ecommerce small enterprise credit appraisal model respectively two groups of samples to be marked; And rule layer index and TOP SCORES result done paired sample t check, analyze the difference that two groups of small enterprises exist on rule layer index and PTS.According to assay; P＜0.01 of operator's quality, online transaction feedback information, Internet business environmental baseline, capital strength, continuity and public 5 indexs of letter record and PTS; There were significant differences on 0.01 level in explanation; The P of management ability index＜0.05 explains that there were significant differences on 0.05 level.So all there is significant difference in the mean value of 6 rule layer indexs and PTS.
Above reliability and reliability and the validity of validity assay explanation the inventive method aspect credit evaluation are all very good.
Description of drawings
Fig. 1 is network market operator's credit appraisal index system of the present invention.
Fig. 2 is the score of certain city Development Co., Ltd each index in the credit appraisal index system that calculates among the embodiment.
Fig. 3 is the credit appraisal report of certain city Development Co., Ltd among the embodiment.
Embodiment
Specify the present invention below in conjunction with embodiment and accompanying drawing, but the present invention is not limited to this.
A kind of network market operator's credit assessment method may further comprise the steps:
(1) confirm network market operator credit appraisal index system, as shown in Figure 1, constitute by destination layer, rule layer and factor layer.Wherein, Destination layer is a network market operator credit appraisal index; Rule layer comprises operator's quality (Character) index, management ability (Control) index, online transaction feedback information (Consumption) index, Internet business environmental baseline (Condition) index, capital strength (Capital) index, continuity and public letter record (Continuity) index; Factor layer is that each the subordinate's index by above-mentioned rule layer constitutes, and is specially: operator's quality index is made up of nature person's credit card, legal person's credit rating, charitable activity and honor, educational background and professional standards, each index of political affiliation; The management ability index by the operation life of abiding by the law continuously, credit evaluation result over the years, be engaged in Internet business headcount, Network employee wage is counted each index and is constituted; Online transaction feedback information index is made up of transaction sequential statistics, dealing money sequential statistics, each index of transaction results character property evaluation; Internet business environmental baseline index is made up of online sign, network trading market establishment time, management place, each index of annual test situation over the years; Capital strength's index is made up of registered capital, asset-liability ratio, number of stock turnover, the accounts receivable number of turnover, each index of sales revenue rate of growth; Continuity and public letter record index are made up of industrial and commercial credit rating, bank credit grading, black/gray list, criminal penalty information, tax credit rating, each index of customs's credit rating.Above-mentioned network market operator's credit appraisal index system has contained 6 aspects, is provided with to amount to 27 investigation dimensions.
(2) obtain data from network trading market, bank and State Administration for Industry & Commerce; And mate through unique sign of network market operator, obtain the data of each network market operator each index in above-mentioned network market operator's credit appraisal index system; The data of each index all are divided into 5 grades, and like this, each data in above-mentioned network market operator's credit appraisal index system just can be mapped in [0,5] interval.
(3) utilize analytical hierarchy process (AHP method) to confirm the weight of each index in the index system:
(1) confirm the weight of each index of rule layer in the index system:
" 1～9 ratio scale method " table that adopts TLSatty to propose compares according to the relative importance of each index in the rule layer for destination layer in twos, and the structure rule layer is judgment matrix in twos; Specific as follows:
Destination layer is the last layer of rule layer, and each index (operator's quality, management ability, online transaction feedback information, Internet business environmental baseline, capital strength, continuity and public letter record) that is in together in the rule layer is designated as B
1, B
2..., B
n(the n value is 6 here), these indexs and destination layer network market operator credit appraisal index (are designated as A
k) related, then the rule layer judgment matrix can be expressed as:
Wherein, b
IjBe meant destination layer network market operator credit appraisal index A
k, B
iIndex is to B
jThe numeric representation of the relative importance of index; b
Ij＞0, press " 1～9 ratio scale method " table value that TLSatty proposes according to expert's questionnaire; b
Ii=1, b
Ij=1/b
Ji, i, j=1,2 ..., n;
In " 1～9 ratio scale method " table that TLSatty proposes, the scale implication is as shown in table 1:
The implication of table 1 1～9 scale during judgment matrix makes up in twos
Calculate the n th root of each row all elements product in the above-mentioned rule layer judgment matrix, that is:
Wherein, b
IjElement for the capable j row of i in the judgment matrix in twos;
Do normalization to these root vectors of trying to achieve again and handle, obtain i characteristic vector W
i, computing formula is:
Thereby, obtain characteristic vector W=[w
1, w
2..., w
n]
T, be the weight of each index of rule layer.
Usually, in order to check the consistance of judgment matrix in twos, need to calculate its coincident indicator C.I., order
Wherein, λ max is the n rank eigenvalue of maximum of judgment matrix in twos, can calculate through following formula to obtain:
In the formula, B is judgment matrix, and Wi is an i proper vector, and W is a proper vector.
When C.I.=0, judgment matrix has crash consistency in twos, is regarded as having satisfied consistance; C.I. big more, the consistance of judgment matrix is just poor in twos.In order to check judgment matrix whether to have satisfied consistance, need C.I. and mean random coincident indicator R.I. be compared.In analytical hierarchy process (AHP method), the value of R.I. is seen table 2.
Table 2 mean random coincident indicator RI value
| n | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 |
| R.I. | 0 | 0 | 0.58 | 0.90 | 1.12 | 1.24 | 1.32 | 1.41 | 1.45 | 1.49 |
Make:
wherein; C.R. be Consistency Ratio at random, R.I. is the mean random coincident indicator with the C.I. same order.When C.R.＜0.1, can think that judgment matrix has satisfied consistance in twos.Otherwise, just must readjust the element in the judgment matrix in twos, till judgment matrix has satisfied consistance in twos.When judgment matrix had satisfied consistance in twos, its characteristic of correspondence vector could be as the weight of each index of rule layer.
(2) confirm the weight of each index of factor layer in the index system:
Adopt the method identical to confirm to comprise the weight of each index of factor layer in the index system with (one):
(i) adopt identical method to confirm six factor layer judgment matrix in twos, be respectively:
Compare the factor layer first of being constructed judgment matrix in twos based on nature person's credit card, legal person's credit rating, charitable activity and honor, educational background and professional standards, each index of political affiliation in twos for the relative importance of operator's quality index;
Based on the operation life of abiding by the law continuously, credit evaluation result over the years, be engaged in Internet business headcount, Network employee wage and count each index and compare the factor layer second of being constructed judgment matrix in twos in twos for the relative importance of management ability index;
Compare the factor layer the 3rd of being constructed judgment matrix in twos based on transaction sequential statistics, dealing money sequential statistics, each index of transaction results character property evaluation in twos for the relative importance of online transaction feedback information index;
Found time, management place, each index of annual test situation over the years based on online sign, network trading market and compare the factor layer the 4th of being constructed judgment matrix in twos in twos for the relative importance of Internet business environmental baseline index;
Compare the factor layer the 5th of being constructed judgment matrix in twos based on registered capital, asset-liability ratio, number of stock turnover, the accounts receivable number of turnover, each index of sales revenue rate of growth in twos for the relative importance of capital strength's index;
Compare the factor layer the 6th of being constructed judgment matrix in twos based on the grading of industrial and commercial credit rating, bank credit, black/gray list, criminal penalty information, tax credit rating, each index of customs's credit rating in twos for the relative importance of continuity and public letter record index;
(ii) adopt the method identical to calculate above-mentioned each proper vector of judgment matrix in twos, the weight of each index of acquisition factor layer with (one):
Above-mentioned factor layer first is the n th root of each row all elements product in the judgment matrix in twos; And carry out normalization respectively and handle; Obtain its characteristic vector W, thereby obtain the weight of nature person's credit card, legal person's credit rating, charitable activity and honor, educational background and professional standards, each index of political affiliation;
Above-mentioned factor layer second is the n th root of each row all elements product in the judgment matrix in twos; And carry out normalization respectively and handle; Obtain its characteristic vector W, thus the operation life of being abide by the law continuously, credit evaluation result over the years, be engaged in Internet business headcount, Network employee wage is counted each index;
Above-mentioned factor layer the 3rd is the n th root of each row all elements product in the judgment matrix in twos; And carry out normalization respectively and handle; Obtain its characteristic vector W, thereby obtain the weight that sequential is added up, the dealing money sequential is added up, the transaction results character property is estimated each index of concluding the business;
Above-mentioned factor layer the 4th is the n th root of each row all elements product in the judgment matrix in twos; And carry out normalization respectively and handle; Obtain its characteristic vector W, time, management place, each index of annual test situation over the years are founded in sign, network trading market on the net thereby get;
Above-mentioned factor layer the 5th is the n th root of each row all elements product in the judgment matrix in twos; And carry out normalization respectively and handle; Obtain its characteristic vector W, thereby obtain the weight of registered capital, asset-liability ratio, number of stock turnover, the accounts receivable number of turnover, each index of sales revenue rate of growth;
Above-mentioned factor layer the 6th is the n th root of each row all elements product in the judgment matrix in twos; And carry out normalization respectively and handle; Obtain its characteristic vector W, thereby obtain the weight of the grading of industrial and commercial credit rating, bank credit, black/gray list, criminal penalty information, tax credit rating, customs's each index of credit rating;
(iii) adopt the above-mentioned respectively consistance of judgment matrix in twos of the method check identical with (one); That is: calculate above-mentioned each eigenvalue of maximum λ max of judgment matrix in twos respectively; And calculate coincident indicator C.I. and Consistency Ratio C.R. at random in view of the above; When C.I.=0, judgment matrix has crash consistency in twos, is regarded as having satisfied consistance; When C.I. is not 0, need judge with reference to C.R., when C.R.＜0.1, can think that judgment matrix has satisfied consistance in twos.Otherwise, just must readjust the element in the judgment matrix in twos, till judgment matrix has satisfied consistance in twos.When judgment matrix had satisfied consistance in twos, its characteristic of correspondence vector could be as the weight of each index of factor layer.
(4) network market operator is carried out credit comprehensive evaluation:
According to the data of each index in the step (2) and the weight of each index in the step (3); Calculate the score of each index in described network market operator's credit appraisal index system; According between the scoring area that each score value dropped on; Its concrete implication is carried out semantic interpretation, and finally provide a credit evaluation report.
For the present invention is described better, will be that example specifies above-mentioned credit assessment method certain city Development Co., Ltd is carried out credit appraisal below:
(1) confirm certain city Development Co., Ltd credit appraisal index system, as shown in Figure 1, detailed content details in above-mentioned steps (1), no longer repeats at this;
(2) obtain data from Alibaba's link integrity data, river province Construction Bank Internet bank's customer data and Zhejiang Province's industrial and commercial registration with annual test information; And mate through unique sign of network market operator, obtain the data of each network market operator each index in above-mentioned network market operator's credit appraisal index system; The data of each index all are divided into 5 grades, and like this, each data in above-mentioned network market operator's credit appraisal index system just can be mapped in [0,5] interval.
(3) utilize the AHP method to confirm the weight of each index in the index system:
(i) confirm the rule layer judgment matrix according to each index (operator's quality, management ability, online transaction feedback information, Internet business environmental baseline, capital strength, continuity and public letter record) that is in together in the rule layer of listing in the following table for the comparative result in twos of the relative importance of destination layer network market operator credit appraisal index.
Confirm factor layer first judgment matrix in twos according to nature person's credit card of listing in the following table, legal person's credit rating, charitable activity and honor, educational background and professional standards, each index of political affiliation for the result that the relative importance of operator's quality index compares in twos.
According to the operation life of abiding by the law continuously of listing in the following table, credit evaluation result over the years, be engaged in Internet business headcount, Network employee wage and count each index and confirm factor layer second judgment matrix in twos for the result that the relative importance of management ability index compares in twos.
Confirm factor layer the 3rd judgment matrix in twos according to the transaction sequential listed in following table statistics, dealing money sequential statistics, each index of transaction results character property evaluation for the result that the relative importance of online transaction feedback information index compares in twos.
| Transaction sequential statistics | Dealing money sequential statistics | The transaction results character property is estimated | |
| Transaction sequential statistics | 1 | 1/2 | 1/3 |
| Dealing money sequential statistics | 2 | 1 | 3 |
| The transaction results character property is estimated | 3 | 1/3 | 1 |
Found time, management place, each index of annual test situation over the years according to the online sign of listing in the following table, network trading market and confirm factor layer the 4th judgment matrix in twos for the result that the relative importance of Internet business environmental baseline index compares in twos.
| The online sign | The time is founded in network trading market | The management place | Annual test situation over the years | |
| The online sign | 1 | 2 | 5 | 3 |
| The time is founded in network trading market | 1/2 | 1 | 1/3 | 2 |
| The management place | 1/5 | 3 | 1 | 6 |
| Annual test situation over the years | 1/3 | 1/2 | 1/6 | 1 |
The result who compares in twos according to the relative importance of the registered capital of listing in the following table, asset-liability ratio, number of stock turnover, the accounts receivable number of turnover, each index of sales revenue rate of growth confirms factor layer the 5th judgment matrix in twos.
Confirm factor layer the 6th judgment matrix in twos according to the grading of the industrial and commercial credit rating of listing in the following table, bank credit, black/gray list, criminal penalty information, tax credit rating, each index of customs's credit rating for the result that the relative importance of continuity and public letter record index compares in twos.
(ii) calculate above-mentioned each proper vector of judgment matrix in twos respectively, obtain the weight of each index in the index system, tabulation as follows:
(iii) calculate above-mentioned each coincident indicator C.I. and Consistency Ratio C.R. at random of judgment matrix in twos respectively, finds above-mentioned respectively in twos judgment matrix have satisfied consistance, the weight of each index that the employing aforementioned calculation obtains.
(4) network market operator is carried out credit comprehensive evaluation:
According to the data of each index in the step (2) and the weight of each index in the step (3), calculate the score of each index in the described credit appraisal index system, as shown in Figure 2.According between the scoring area that each score value dropped on, its concrete implication is carried out semantic interpretation, and finally provide a credit evaluation report, as shown in Figure 3.
In addition, also The above results has been carried out the cross check.According to the traditional bank rating result; 100 tame electronics commercial affairs small enterprise sample can be divided into medium and two groups of above, low middlings; Adopt ecommerce small enterprise credit appraisal model respectively two groups of samples to be marked; And rule layer index and TOP SCORES result done paired sample t check, analyze the difference that two groups of small enterprises exist on rule layer index and PTS.
According to assay; P＜0.01 of operator's quality (being called for short individual quality in the table), online transaction feedback information (being called for short online transaction in the table), Internet business environmental baseline (being called for short business environment in the table), capital strength, continuity and public letter record (being called for short public letter record in the table) and PTS; There were significant differences on 0.01 level in explanation; The P of management ability index＜0.05 explains that there were significant differences on 0.05 level.So all there is significant difference in the mean value of 6 rule layer indexs and PTS, explains that reliability and the validity of the inventive method aspect credit evaluation is all very good.
The result of paired t-test sees following table for details
Claims (4)
1. a network market operator credit assessment method is characterized in that, may further comprise the steps:
(1) confirms network market operator credit appraisal index system;
Described network market operator's credit appraisal index system is made up of destination layer, rule layer and factor layer; Wherein, Described destination layer is a network market operator credit appraisal index; Described rule layer comprises operator's quality index, management ability index, online transaction feedback information index, Internet business environmental baseline index, capital strength's index, continuity and public letter record index; Described factor layer is that each the subordinate's index by said rule layer constitutes; Wherein, Said operator's quality index is made up of nature person's credit card, legal person's credit rating, charitable activity and honor, educational background and professional standards, each index of political affiliation; Said management ability index by the operation life of abiding by the law continuously, credit evaluation result over the years, be engaged in Internet business headcount, Network employee wage is counted each index and is constituted; Said online transaction feedback information index is made up of transaction sequential statistics, dealing money sequential statistics, each index of transaction results character property evaluation; Said Internet business environmental baseline index is founded time, management place, each index of annual test situation over the years by online sign, network trading market and is constituted, and said capital strength index is made up of registered capital, asset-liability ratio, number of stock turnover, the accounts receivable number of turnover, each index of sales revenue rate of growth, and said continuity and public letter record index are made up of industrial and commercial credit rating, bank credit grading, black/gray list, criminal penalty information, tax credit rating, each index of customs's credit rating;
(2) obtain data from network trading market, bank and State Administration for Industry & Commerce; And mate through unique sign of network market operator, obtain the data of each network market operator each index in described network market operator's credit appraisal index system; And the data of each index all are divided into 5 grades, like this, each data in described network market operator's credit appraisal index system just can be mapped in [0,5] interval;
(3) utilize analytical hierarchy process to confirm the weight of each index in the index system:
(3.1) compare in twos according to the relative importance of each index in the same level, construct judgment matrix in twos respectively for last layer correlations index;
(3.2) calculate the proper vector of said judgment matrix in twos, obtain the weight of each index in the index system;
(3.3) consistance of the said judgment matrix in twos of check judges to have satisfied consistance in twos as said, then adopts the weight of each index that step (3.2) calculated; Otherwise, return step (3.1) and readjust the element in the judgment matrix in twos, and repeating step (3.2)～(3.3), till judgment matrix has satisfied consistance in twos;
(4) network market operator is carried out credit comprehensive evaluation:
According to the data of each index in the step (2) and the weight of each index in the step (3); Calculate the score of each index in described network market operator's credit appraisal index system; According between the scoring area that each score value dropped on; Its concrete implication is carried out semantic interpretation, and finally provide a credit evaluation report.
2. network market operator's credit assessment method as claimed in claim 1 is characterized in that, step (3.1) comprising:
Compare the rule layer of being constructed judgment matrix in twos based on operator's quality of rule layer, management ability, online transaction feedback information, Internet business environmental baseline, capital strength, continuity and public each index of letter record in twos for the relative importance of destination layer network market operator credit appraisal index;
Compare the factor layer first of being constructed judgment matrix in twos based on nature person's credit card, legal person's credit rating, charitable activity and honor, educational background and professional standards, each index of political affiliation in twos for the relative importance of operator's quality index;
Based on the operation life of abiding by the law continuously, credit evaluation result over the years, be engaged in Internet business headcount, Network employee wage and count each index and compare the factor layer second of being constructed judgment matrix in twos in twos for the relative importance of management ability index;
Compare the factor layer the 3rd of being constructed judgment matrix in twos based on transaction sequential statistics, dealing money sequential statistics, each index of transaction results character property evaluation in twos for the relative importance of online transaction feedback information index;
Found time, management place, each index of annual test situation over the years based on online sign, network trading market and compare the factor layer the 4th of being constructed judgment matrix in twos in twos for the relative importance of Internet business environmental baseline index;
Compare the factor layer the 5th of being constructed judgment matrix in twos based on registered capital, asset-liability ratio, number of stock turnover, the accounts receivable number of turnover, each index of sales revenue rate of growth in twos for the relative importance of capital strength's index;
With compare the factor layer the 6th of being constructed judgment matrix in twos based on the grading of industrial and commercial credit rating, bank credit, black/gray list, criminal penalty information, tax credit rating, each index of customs's credit rating in twos for the relative importance of continuity and public letter record index.
3. network market operator's credit assessment method as claimed in claim 1 is characterized in that, step (3.2) comprising:
Tie up judgment matrix in twos for said n, calculate the n th root of each row all elements product in the judgment matrix in twos, that is:
Wherein, b
IjElement for the capable j row of i in the judgment matrix in twos;
Do normalization to these root vectors of trying to achieve again and handle, obtain i characteristic vector W
i, computing formula is:
Thereby, obtain characteristic vector W=[w
1, w
2..., w
n]
T, be the weight of each index.
4. like claim 1 or 3 described network market operator's credit assessment methods; It is characterized in that; In the said step (3.3); Check the consistance of said judgment matrix in twos, be through calculating said judgment matrix in twos coincident indicator C.I. and at random the value of Consistency Ratio C.R. judge:
When C.I.=0, judgment matrix has crash consistency in twos, is regarded as having satisfied consistance; When C.I. is not 0, judge according to the value of C.R., when C.R.＜0.1, think that judgment matrix has satisfied consistance in twos, otherwise, think that judgment matrix does not have satisfied consistance in twos;
Wherein, described coincident indicator C.I. calculates through following formula:
Wherein, λ max is the n rank eigenvalue of maximum of judgment matrix in twos,
B is said n rank judgment matrixs in twos, W
iBe i proper vector, W is a proper vector;
The described C.R. of Consistency Ratio at random calculates through following formula:
R.I. be mean random coincident indicator in the analytical hierarchy process.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011104460532A CN102436622A (en) | 2011-12-28 | 2011-12-28 | Method for evaluating network market operator credit status |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011104460532A CN102436622A (en) | 2011-12-28 | 2011-12-28 | Method for evaluating network market operator credit status |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102436622A true CN102436622A (en) | 2012-05-02 |
Family
ID=45984670
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2011104460532A Pending CN102436622A (en) | 2011-12-28 | 2011-12-28 | Method for evaluating network market operator credit status |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102436622A (en) |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103065271A (en) * | 2013-01-23 | 2013-04-24 | 广东电网公司东莞供电局 | Power supply reliability and customer satisfaction quantitative relation model establishment method |
| CN103366090A (en) * | 2013-07-11 | 2013-10-23 | 西安交通大学 | Index weight assessment method based on section rating of experts |
| CN103514566A (en) * | 2013-10-15 | 2014-01-15 | 国家电网公司 | Risk control system and method |
| CN104123656A (en) * | 2014-07-14 | 2014-10-29 | 广东省标准化研究院 | Credit evaluation method based on AHP (analytic hierarchy process) |
| CN104508623A (en) * | 2012-05-30 | 2015-04-08 | 邓白氏公司 | Credit behavior network mapping |
| CN104680297A (en) * | 2013-12-03 | 2015-06-03 | 航天信息股份有限公司 | Evaluation system and evaluation method for credit rating |
| CN105678602A (en) * | 2015-12-31 | 2016-06-15 | 天津迈沃医药技术有限公司 | Enterprise credit evaluation method and system used for self-diagnosis platform |
| CN106997472A (en) * | 2016-01-25 | 2017-08-01 | 阿里巴巴集团控股有限公司 | User characteristics sorting technique, user credit appraisal procedure and the device of user credit model |
| CN107563645A (en) * | 2017-09-04 | 2018-01-09 | 杭州云算信达数据技术有限公司 | A kind of Financial Risk Analysis method based on big data |
| CN108446848A (en) * | 2018-03-21 | 2018-08-24 | 北京理工大学 | Individual networks awareness of safety scalar quantization evaluation method |
| CN109993409A (en) * | 2019-03-01 | 2019-07-09 | 北京农业智能装备技术研究中心 | Operator of farm machinery's level of operation evaluation method and device |
| CN110264343A (en) * | 2019-06-21 | 2019-09-20 | 山东浪潮商用系统有限公司 | A kind of credit estimation method based on tax big data analysis |
| CN110826835A (en) * | 2018-08-14 | 2020-02-21 | 广州中国科学院先进技术研究所 | Glove-based acupuncture training method, system, platform and storage medium |
| CN111541702A (en) * | 2020-04-27 | 2020-08-14 | 北京天际友盟信息技术有限公司 | Network threat security detection method and device |
| CN112836931A (en) * | 2020-12-31 | 2021-05-25 | 广州智能科技发展有限公司 | Method for generating multidimensional personal credit evaluation model |
| CN112926833A (en) * | 2021-01-28 | 2021-06-08 | 北京安九信息技术有限公司 | Detection method and detection system for health condition of enterprise |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101493913A (en) * | 2008-01-23 | 2009-07-29 | 阿里巴巴集团控股有限公司 | Method and system for assessing user credit in internet |
| CN101685519A (en) * | 2008-09-22 | 2010-03-31 | 浙江大学 | Credit evaluation method and credit evaluation system |
-
2011
- 2011-12-28 CN CN2011104460532A patent/CN102436622A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101493913A (en) * | 2008-01-23 | 2009-07-29 | 阿里巴巴集团控股有限公司 | Method and system for assessing user credit in internet |
| CN101685519A (en) * | 2008-09-22 | 2010-03-31 | 浙江大学 | Credit evaluation method and credit evaluation system |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104508623A (en) * | 2012-05-30 | 2015-04-08 | 邓白氏公司 | Credit behavior network mapping |
| CN104508623B (en) * | 2012-05-30 | 2016-09-21 | 邓白氏公司 | Behavior of credit network mapping |
| CN103065271B (en) * | 2013-01-23 | 2018-09-07 | 广东电网公司东莞供电局 | A kind of power supply reliability and customer satisfaction relationship quantitative model method for building up |
| CN103065271A (en) * | 2013-01-23 | 2013-04-24 | 广东电网公司东莞供电局 | Power supply reliability and customer satisfaction quantitative relation model establishment method |
| CN103366090A (en) * | 2013-07-11 | 2013-10-23 | 西安交通大学 | Index weight assessment method based on section rating of experts |
| CN103514566A (en) * | 2013-10-15 | 2014-01-15 | 国家电网公司 | Risk control system and method |
| CN104680297A (en) * | 2013-12-03 | 2015-06-03 | 航天信息股份有限公司 | Evaluation system and evaluation method for credit rating |
| CN104123656A (en) * | 2014-07-14 | 2014-10-29 | 广东省标准化研究院 | Credit evaluation method based on AHP (analytic hierarchy process) |
| CN105678602A (en) * | 2015-12-31 | 2016-06-15 | 天津迈沃医药技术有限公司 | Enterprise credit evaluation method and system used for self-diagnosis platform |
| CN106997472A (en) * | 2016-01-25 | 2017-08-01 | 阿里巴巴集团控股有限公司 | User characteristics sorting technique, user credit appraisal procedure and the device of user credit model |
| CN106997472B (en) * | 2016-01-25 | 2020-06-02 | 阿里巴巴集团控股有限公司 | User feature classification method of user credit model, user credit evaluation method and device |
| CN107563645A (en) * | 2017-09-04 | 2018-01-09 | 杭州云算信达数据技术有限公司 | A kind of Financial Risk Analysis method based on big data |
| CN108446848A (en) * | 2018-03-21 | 2018-08-24 | 北京理工大学 | Individual networks awareness of safety scalar quantization evaluation method |
| CN110826835A (en) * | 2018-08-14 | 2020-02-21 | 广州中国科学院先进技术研究所 | Glove-based acupuncture training method, system, platform and storage medium |
| CN109993409A (en) * | 2019-03-01 | 2019-07-09 | 北京农业智能装备技术研究中心 | Operator of farm machinery's level of operation evaluation method and device |
| CN110264343A (en) * | 2019-06-21 | 2019-09-20 | 山东浪潮商用系统有限公司 | A kind of credit estimation method based on tax big data analysis |
| CN111541702A (en) * | 2020-04-27 | 2020-08-14 | 北京天际友盟信息技术有限公司 | Network threat security detection method and device |
| CN112836931A (en) * | 2020-12-31 | 2021-05-25 | 广州智能科技发展有限公司 | Method for generating multidimensional personal credit evaluation model |
| CN112926833A (en) * | 2021-01-28 | 2021-06-08 | 北京安九信息技术有限公司 | Detection method and detection system for health condition of enterprise |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102436622A (en) | Method for evaluating network market operator credit status | |
| Amiram et al. | Financial statement errors: Evidence from the distributional properties of financial statement numbers | |
| Mehralian et al. | Intellectual capital and corporate performance in Iranian pharmaceutical industry | |
| Price III et al. | Detecting and predicting accounting irregularities: A comparison of commercial and academic risk measures | |
| Hellerstein et al. | Market forces and sex discrimination | |
| Kim et al. | Ratio analysis for the hospitality industry: A cross sector comparison of financial trends in the lodging, restaurant, airline, and amusement sectors | |
| Kamal et al. | Intellectual capital and firm performance of commercial banks in Malaysia | |
| Brau | Do banks price owner–manager agency costs? An examination of small business borrowing | |
| Akçomak et al. | Measuring and interpreting trends in the division of labour in the Netherlands | |
| Iranmahd et al. | The effect of intellectual capital on cost of finance and firm value | |
| Podgursky et al. | Student loan defaults and enrollment persistence | |
| Kurtulus et al. | An empirical analysis of the relationship between employee ownership and employment stability in the US: 1999–2011 | |
| Eshleman et al. | Do seasoned industry specialists provide higher audit quality? A re-examination | |
| Nwankwo et al. | Effect of cooperative type and age on profit performance: A study of cooperative societies in Awka North LGA in Anambra State, Nigeria | |
| Blaum et al. | Firm size and the intensive margin of import demand | |
| Baldwin et al. | The importance of entry to Canadian manufacturing with an appendix on measurement issues | |
| Tadić et al. | Defining human resources “bundles” and its’ correlation with companies’ financial performances | |
| Singh et al. | Does financial performance depend on hotel size? Analysis of the financial profile of the US lodging industry. | |
| Ding et al. | Effects of corporate social performance on corporate financial performance: A two-sector analysis between the US hospitality and manufacturing companies | |
| Amiri et al. | Diversity in the board of directors and the social sustainability pillar of the firm: Evidence from countries with high environmental, social, and governance scores | |
| CN113159634A (en) | Financial product management method and device and electronic equipment | |
| Rigó | Estimating union-non-union wage differential in hungary | |
| Rogers et al. | Firm performance and investment in R&D and intellectual property | |
| Negulescu | The management process of the enterprise valuation | |
| YILDIRIM et al. | Comparison of the Companies on the BIST Sustainability Index with Other Listed Companies in the Context of Earnings Manipulation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20120502 |