CN108009751A - A kind of power grid enterprises' asset quality diagnosis and method for early warning - Google Patents
A kind of power grid enterprises' asset quality diagnosis and method for early warning Download PDFInfo
- Publication number
- CN108009751A CN108009751A CN201711426676.7A CN201711426676A CN108009751A CN 108009751 A CN108009751 A CN 108009751A CN 201711426676 A CN201711426676 A CN 201711426676A CN 108009751 A CN108009751 A CN 108009751A
- Authority
- CN
- China
- Prior art keywords
- power grid
- asset quality
- grid enterprises
- index
- asset
- 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
- 238000003745 diagnosis Methods 0.000 title claims abstract description 45
- 238000000034 method Methods 0.000 title claims abstract description 41
- 238000007637 random forest analysis Methods 0.000 claims abstract description 24
- 238000004458 analytical method Methods 0.000 claims abstract description 15
- 239000011159 matrix material Substances 0.000 claims description 34
- 238000012545 processing Methods 0.000 claims description 12
- 238000012549 training Methods 0.000 claims description 12
- 238000011156 evaluation Methods 0.000 claims description 10
- 239000000203 mixture Substances 0.000 claims description 5
- 230000008569 process Effects 0.000 abstract description 7
- 238000013459 approach Methods 0.000 abstract description 2
- 238000003066 decision tree Methods 0.000 description 8
- 230000008901 benefit Effects 0.000 description 5
- 238000004422 calculation algorithm Methods 0.000 description 4
- 238000002405 diagnostic procedure Methods 0.000 description 4
- 238000011161 development Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000005070 sampling Methods 0.000 description 3
- 230000007306 turnover Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 230000007812 deficiency Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 206010068052 Mosaicism Diseases 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 239000004744 fabric Substances 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 239000007788 liquid Substances 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000000513 principal component analysis Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 210000003765 sex chromosome Anatomy 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000004083 survival effect Effects 0.000 description 1
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
- 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
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- 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
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Development Economics (AREA)
- Health & Medical Sciences (AREA)
- Educational Administration (AREA)
- Marketing (AREA)
- Entrepreneurship & Innovation (AREA)
- Theoretical Computer Science (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Game Theory and Decision Science (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention relates to a kind of diagnosis of power grid enterprises' asset quality and method for early warning;Including step S1, structure power grid enterprises asset quality diagnosis index system;Step S2, the power grid asset quality comprehensive diagnostic model based on principal component TOPSIS methods is built, asset quality is diagnosed;Step S3, build power grid enterprises' asset quality Early-warning Model and early warning analysis is carried out to asset quality.Reflect the feature of power grid enterprises' assets Comprehensive of the present invention;The complexity of model is greatly reduced on the basis of index internal information gain is ensured, simplifies the calculating process of model;It is proposed power grid enterprises' asset quality Early-warning Model based on random forest, can look-ahead enterprise assets quality level, be effectively prevented from the asset quality problem being likely to occur, analyzed for power grid enterprises asset quality status early warnings and a kind of new approaches are provided.
Description
Technical field
The present invention relates to power grid enterprises' asset quality field, more particularly to a kind of diagnosis of power grid enterprises' asset quality and early warning
Method.
Background technology
The assets of enterprise are the material bases that enterprise depended on and obtained profit for existence, and the quality of enterprise assets quality determines enterprise
The survival state of industry and the complexity for obtaining profit.Enterprise assets quality diagnosis is to Corporate Asset Management business activities performance
Judge, be to ensure that asset quality reaches one of main management measure of good level.Asset quality diagnostic result can be into one
Walk clear and definite Corporate Asset Management responsibility, promote Corporate Asset Management to improving benefit, reduce cost, avoid risk, reduce loss
In-depth development, constantly improve condition of assets, level of improving asset quality.
In recent years, deepening continuously with extra-high voltage grid construction, the asset size of power grid enterprises constantly expand, add outside
Overseas M & A Facing implementation, power grid enterprises are faced with huge asset operation and government pressure.For a long time, power grid enterprises provide
Produce the outstanding feature that has be intensive capital, it is technology-intensive, how efficiently to run huge assets, realize the assets of power grid enterprises
Optimum management, is the key issue for improving power grid enterprises' business performance.Therefore, asset quality level is lifted to control power grid enterprises
Business risk, keep sustainable development to be of great significance.Realize the early warning analysis to power grid asset quality, can be effectively pre-
It is horizontal to survey asset quality, provides to be likely to occur asset quality problem and gives warning in advance, effectively evade asset quality and decline to enterprise
The potential risk that industry is brought.
At present domestic scholars for the research of power grid enterprises asset qualities diagnosis to concentrating on the structure of diagnosis index system,
Diagnostic method and early warning technology are not furtherd investigate.Scholar is proposed from asset structure, assets earning capacity, asset operation effect
Rate, asset risk degree etc. build diagnosis index, based on analytic hierarchy process (AHP) agriculture products weight.The diagnosis side of scholar's research
Method also focuses mostly in expert's subjective judgement, does not take into full account the information that index itself contains.Suitable electric power enterprise is not suggested that
The method of asset quality diagnostic analysis and early warning technology.
The content of the invention
In view of above-mentioned analysis, the present invention is intended to provide a kind of diagnosis of power grid enterprises' asset quality and method for early warning, to
Solve the problems, such as that existing model and method can not carry out Accurate Diagnosis and analysis to power grid enterprises' asset quality.
The purpose of the present invention is mainly achieved through the following technical solutions:
A kind of power grid enterprises' asset quality diagnosis and method for early warning;Comprise the following steps:
Step S1, power grid enterprises' asset quality diagnosis index system is built;
Step S2, build the power grid asset quality comprehensive diagnostic model based on principal component TOPSIS methods, to asset quality into
Row diagnosis;
Step S3, build power grid enterprises' asset quality Early-warning Model and early warning analysis is carried out to asset quality.
Further, the index system includes three assets efficiency, earning capacity and debt paying ability dimension evaluation indexes.
Further, the step 2 includes following sub-step:
Step S201, evaluation index original matrix is constructedWherein, n is power grid enterprises' sample
Number;M is asset quality Characteristic Number, xijFor i-th of power grid enterprises, j-th of characteristic index numerical value, i=1, n, j=
1,···,m;
Step S202, the index in original matrix X is pre-processed to obtain standard observation matrix Y;
Step S203, the coefficient R of standard observation matrix Y is solved;
Step S204, the characteristic root λ of correlation matrix R is solvedj, and characteristic root is sorted from big to small, determine it is main into
Point and principal component number p;
Step S205, the corresponding feature vector of principal component characteristic root and its principal component matrix F are extracted;
Step S206, determine in principal component matrix F, characteristic root λjCorresponding principal component column vector fjTo the distance of ideal point,
The ideal point is the maximum or minimum point of each row;
Step S207, calculated according to the distance to ideal point and be lined up indicated value, and good and bad row is carried out according to indicated value is lined up
Sequence, obtains asset quality diagnostic result.
Further, the pretreatment includes:
1) to evaluation index xijUnification processing is carried out, is converted into large index;
2) nondimensionalization processing is carried out to the index for being converted into large, obtained by standard observation value yijThe standard of composition is seen
Survey matrix Y.
Further, the unification processing includes two ways:
A. the index of minimal type, according to formula x'ij=1-xij, it is converted into large index;
B. for the index of interval type, according to formulaIt is converted into greatly
Type index;Wherein, [q1,q2] it is xijOptimal section, xminAnd xmaxFor xijThe minimum value and maximum of possible value.
Further, the standard observation valueIn formula,And sjThe respectively sample of jth item characteristic index value
This average and sample variance.
Further, the selection of the principal component is with λj> 1 and contribution rate of accumulative totalAs standard.
Further, the distance to ideal point includes fijTo the distance D of Positive ideal pointj +With to Negative ideal point away from
From Dj -;
fijTo the distance of Positive ideal pointWherein, Positive ideal point zj +For column vector fjMaximum
Value;
fijTo the distance of Negative ideal pointWherein;Negative ideal point zj -For column vector fjMinimum
Value;
The queuing indicated value is according to formulaJ=1,2 ..., p is obtained.
Further, the step 3 uses random forest Early-warning Model as power grid enterprises' asset quality Early-warning Model.
Further, step 3 includes following sub-step:
Step S301, the asset quality being calculated according to the principal component matrix F and principal component TOPSIS models diagnoses
As a result classify, data set T of the structure new data set as random forest Early-warning Model;
Step S302, random forest Early-warning Model is trained using data set T;
Step S303, using the random forest Early-warning Model after training to the horizontal identification and classification of asset quality
Step S304, classified according to early warning, take different reply management measures.
The present invention has the beneficial effect that:
The present invention realizes the diagnosis and early warning to power grid enterprises' asset quality.Specifically, from assets efficiency, profit energy
Three dimensions of power and debt paying ability construct the index system suitable for power grid enterprises' asset quality diagnosis, reflect Comprehensive
The features of power grid enterprises' assets;
Herein on basis, the TOPSIS diagnostic methods based on principal component dimensionality reduction degree are established, are ensureing index internal information
The complexity of model is greatly reduced on the basis of gain, simplifies the calculating process of model, is examined for power grid enterprises' asset quality
It is disconnected to propose a kind of new method;
Further, propose power grid enterprises' asset quality Early-warning Model based on random forest, make full use of random forest
Calculating process is simple, suitable for high dimensional feature data set and the advantages of nicety of grading height is not easy over-fitting, generalization ability is high,
Can look-ahead enterprise assets quality level, be effectively prevented from the asset quality problem being likely to occur, be power grid enterprises' assets
Quality state early warning analysis provides a kind of new approaches.
Other features and advantages of the present invention will illustrate in the following description, also, partial become from specification
Obtain it is clear that or being understood by implementing the present invention.The purpose of the present invention and other advantages can be by the explanations write
Specifically noted structure is realized and obtained in book, claims and attached drawing.
Brief description of the drawings
Attached drawing is only used for showing the purpose of specific embodiment, and is not considered as limitation of the present invention, in whole attached drawing
In, identical reference symbol represents identical component.
Fig. 1 is power grid enterprises' asset quality diagnosis and method for early warning flow chart;
Fig. 2 is power grid enterprises' asset quality diagnosis index system.
Embodiment
The preferred embodiment of the present invention is specifically described below in conjunction with the accompanying drawings, wherein, attached drawing forms the application part, and
It is used to explain the principle of the present invention together with embodiments of the present invention.
The specific embodiment of the present invention, discloses a kind of power grid enterprises' asset quality diagnosis and method for early warning;Such as Fig. 1
It is shown, specifically include step:
Step S1, power grid enterprises' asset quality diagnosis index system is built;
Power grid enterprises' asset quality diagnosis index system needs to consider structure, operation, payment of debts and the development of assets comprehensively, dashes forward
Go out the safety, efficiency and benefit of assets, and fully pay close attention to the core asset quality of power grid.Based on considerations above, the present embodiment
Index systems are built from three dimensions such as assets efficiency, earning capacity and debt paying ability, as shown in Figure 2:
Index under assets efficiency dimension has, the turnover of total assets, the velocity of liquid assets, accounts receivable turnover, solid
Determine asset turnover, Unit Assets electricity sales amount, Unit Assets powering area.
Index under earning capacity dimension has, and net assets income ratio, assets return, net profit on sales rate, government capital are protected
It is worth appreciation rate, Unit Assets contribution margin.
Index under debt paying ability dimension has, asset-liability ratio, liquidity ratio, credits ratio, current rate, liability with interest
Ratio.
Step S2, build the power grid asset quality comprehensive diagnostic model based on principal component TOPSIS methods, to asset quality into
Row diagnosis
The composite diagnostic index system built in step S1 from three assets efficiency, earning capacity and debt paying ability dimensions,
The feature of diagnosis object is held to greatest extent.Although each index all reflects the feature of diagnosis object to a certain extent,
It is that the information reflected is inevitably overlapping, it is this overlapping just even more serious with the increase of index quantity.Based on this reason,
It is proposed to introduce principal component analytical method, eliminate the conllinear sex chromosome mosaicism between index, diagnosis index system dimension is reduced, with less
Index reflection diagnoses the feature of object.Since the situation of principal component analysis factor loading negative occurs, comprehensive diagnos function meaning
It is indefinite, therefore ideal point method (TOPSIS) diagnostic method is introduced, make up the deficiency of principal component analytical method.Specifically include following
Sub-step:
Step S201, the evaluation index original matrix of n power grid enterprises is constructedWherein xijFor
I-th of power grid enterprises, j-th of characteristic index numerical value, i=1, n, j=1, m;N is power grid enterprises' sample
Number;M is asset quality Characteristic Number.
Step S202, the index in original matrix X is pre-processed to obtain standard observation matrix Y;
1) to evaluation index xijUnification processing is carried out, is converted into large index;
The evaluation index includes:
Large index:The value of index is the bigger the better;
Minimal type index:The value of index is the smaller the better;
Interval type index:It is best that the value of index preferably falls in some definite section.
The unification processing of the evaluation index includes two ways:
A. the index of minimal type, according to formula x'ij=1-xij, it is converted into large index;
B. for the index of interval type, according to formulaIt is converted into greatly
Type index;Wherein, [q1,q2] it is xijOptimal section, xminAnd xmaxFor xijThe minimum value and maximum of possible value.
2) nondimensionalization processing is carried out to the index for being converted into large, obtained by standard observation value yijThe standard of composition is seen
Survey matrix Y
During comprehensive diagnos, in order to eliminate index due to existing incommensurability between different dimensions or magnitude,
Nondimensionalization processing is carried out to index before the diagnosis:In formula:And sjRespectively jth item characteristic index value
Sample average and sample variance;yijFor the desired value after standard processing.
Step S203, the coefficient R of standard observation matrix Y is solved;
Utilize formulaTo by yijThe matrix Y of composition solves coefficient R;
Step S204, the characteristic root λ of correlation matrix R is solvedj, and characteristic root is sorted from big to small, determine it is main into
Point and principal component number p;
Utilize | R- λ I | ...=0 solves the characteristic root λ of correlation matrix Rj, I is unit matrix;
By characteristic root λjSort from big to small, according to λj> 1 and contribution rate of accumulative totalP (p are chosen as standard
≤ m) a characteristic root is as principal component characteristic root.
Step S205, the corresponding feature vector of principal component characteristic root and its principal component matrix F are extracted;
According to RUj=λjUjSeek characteristic root λjCorresponding feature vector Uj, the principal component vector f of extractionj=YjUj, wherein, j
=1,2 ..., p;Principal component vector fjPrincipal component matrix F is formed, the principal component matrix F is n × p matrix.
Step S206, determine in principal component matrix, characteristic root λjCorresponding principal component column vector fjTo the distance of ideal point,
The ideal point is the maximum or minimum point of each row;
The distance to ideal point includes fijTo the distance D of Positive ideal pointj +With the distance D to Negative ideal pointj -;
fijTo the distance of Positive ideal pointWherein, Positive ideal point zj +For column vector fjMaximum
Value;
fijTo the distance of Negative ideal pointWherein;Negative ideal point zj -For column vector fjMinimum
Value.
Step S207, calculated according to the distance to ideal point and be lined up indicated value, and good and bad row is carried out according to indicated value is lined up
Sequence, obtains asset quality diagnostic result
Calculate the queuing indicated value of each diagnosis objectJ=1,2 ..., p;
The queuing indicated value emphasizes the distance with minus ideal result, by CjIt is descending that alternative objects are carried out with good and bad row
Sequence, obtains asset quality diagnostic result.
Sample is ranked up according to order from big to small according to indicated value is lined up, according to ranking results according to normal state point
Cloth 1:4:The ratio of 1 ((- ∞, μ-σ), (μ-σ, μ+σ), (μ+σ ,+∞), wherein μ are the average of normal distribution, and σ is standard deviation)
Sample asset quality is divided into healthy, good, general three classes, is represented respectively with a, b, c.
Step S3, build power grid enterprises' asset quality Early-warning Model based on random forest and early warning point is carried out to asset quality
Analysis.
Random forest (RF) algorithm belongs to a kind of algorithm of machine learning in data mining, which existed by Breiman
On the basis of CART sorting algorithms, the thought of Stochastic Decision-making forest is used for reference, during the classification tree that machine learning is produced, is utilized
To the random combine of row variable and row variable data, generate random assortment tree, be then aggregated into forest by these trees, formed with
Machine forest.Random forest improves precision of prediction on the premise of operand does not significantly improve, and is to aim at higher-dimension small sample number
According to classification and return a kind of algorithm being designed.
The sub-step of power grid enterprises' asset quality Early-warning Model based on random forest is as follows.
Step S301, the assets that principal component matrix F and principal component TOPSIS models in step S206 are calculated
Quality diagnosis result is classified, data set of the structure new data set as random forest Early-warning Model;
Step S302, training random forest Early-warning Model
The process of training random forest is exactly the process of trained each decision tree, since the training of each decision tree is mutual
Independent, therefore the training of random forest can be realized by parallel processing, this will greatly improve the efficiency of generation model.
Random forest Early-warning Model is by decision tree { h (F, θk), k=1,2 ..., n } set, F is input vector, θkTo be only
The vertical random vector with distribution.Assuming that training set is T={ (fi,ri), fiFor feature vector, riFor class vector.
(1) Bootstrap sampling (Bootstrap sampling), generation training sample set T are carried out to training set Ti, each sample set structure
Make a corresponding decision tree.Bootstrap sampling refers to be concentrated with randomly selecting with putting back to and original sample collection number from original sample
Measure the training sample set of formed objects.
(2) M composition characteristic subset is randomly selected from P feature selected by step S204, usual M takes
(3) according to Gini indexes, optimal disruptive features are selected from M feature as split vertexes.The meter of Gini indexes
Calculation method is as follows.
Gini=p1(1-p1)+p2(1-p2)+p3(1-p3)
Wherein, p1、p2、p3Belong to the probability of a, b, c three classes for sample.Node purity is lower, and Gini values are bigger.
If sample set TiAccording to some feature MlIt is divided into Ti1And Ti2Two parts, then in feature MlUnder conditions of,
Set TiGini indexes be
(4) division grows to maximum until setting, and circulates (1)-(3), until establishing k decision tree, the collection of tree is combined into { hi}。
(5) sample to be tested classification is chosen in a vote by decision tree set.
Wherein, the classification results of H (f) random forests, hi(f) it is single decision tree classification as a result, R is class object, I
() is indicative function.
Step S303, using the random forest Early-warning Model after training to the horizontal identification and classification of asset quality
The power grid enterprises' asset quality diagnosis index and feature vector U of early warning analysis are carried out as neededjCalculate principal component
Matrix;The principal component sub-matrix for obtaining sample to be sorted is input in the random forest Early-warning Model that training obtains, can obtained
To the output result of model.
Step S304, classified according to early warning, take different reply management measures.
If early warning classification results are outstanding, show that power grid enterprises' asset quality health, risk are smaller, maintain operation at present
State.If it is categorized as well, showing that utility power grid asset quality is not up to the state of optimizing management, enterprise is excellent according to self-growth
Change asset portfolio, realize assets optimum state.If classification results are general, show that enterprise assets quality level is not good enough, enterprise needs
Assets are prepared with management again, makes up to Optimal State.
In conclusion an embodiment of the present invention provides a kind of diagnosis of power grid enterprises' asset quality and method for early warning.This method
Construct and a set of be suitable for power grid enterprises' asset quality diagnostic analysis index system, it is proposed that based on principal component TOPSIS methods
Power grid enterprises' asset quality diagnostic model, effectively overcomes the subjectivity of expert diagnosis model, is taking into full account what index contained
Model dimension, simplified model complexity are reduced while internal information gain, and overcomes master by means of TOPSIS diagnostic methods
The deficiency of Component Model, method is provided for power grid enterprises' asset quality diagnosis.On this basis, propose a kind of based on random gloomy
Power grid enterprises' asset quality method for early warning of woods, efficiently solves the problems, such as the over-fitting of decision tree, is particularly suitable for the higher-dimension number of degrees
Analyzed according to the discriminant classification of collection, this method has effectively filled up the blank of power grid enterprises' asset quality early warning analysis.In addition, this hair
Bright calculating thinking is simple, and calculating process can be realized by software, easy to operate, can meet to look forward to power grid in maximum possible
The requirement of diagnosis, analysis and the early warning of industry asset quality, to be provided for policymaker and manager with reference to support.
It will be understood by those skilled in the art that realizing all or part of flow of above-described embodiment method, meter can be passed through
Calculation machine program instructs relevant hardware to complete, and the program can be stored in computer-readable recording medium.Wherein, institute
Computer-readable recording medium is stated as disk, CD, read-only memory or random access memory etc..
The foregoing is only a preferred embodiment of the present invention, but protection scope of the present invention be not limited thereto,
Any one skilled in the art the invention discloses technical scope in, the change or replacement that can readily occur in,
It should be covered by the protection scope of the present invention.
Claims (10)
1. a kind of power grid enterprises' asset quality diagnosis and method for early warning;It is characterised in that it includes following steps:
Step S1, power grid enterprises' asset quality diagnosis index system is built;
Step S2, the power grid asset quality comprehensive diagnostic model based on principal component TOPSIS methods is built, asset quality is examined
It is disconnected;
Step S3, build power grid enterprises' asset quality Early-warning Model and early warning analysis is carried out to asset quality.
2. power grid enterprises' asset quality diagnosis according to claim 1 and method for early warning, it is characterised in that the index body
System includes three assets efficiency, earning capacity and debt paying ability dimension evaluation indexes.
3. power grid enterprises' asset quality diagnosis according to claim 1 and method for early warning, it is characterised in that the step 2
Including following sub-step:
Step S201, evaluation index original matrix is constructedWherein, n is power grid enterprises' number of samples;
M is asset quality Characteristic Number, xijFor i-th of power grid enterprises, j-th of characteristic index numerical value, i=1, n, j=
1,···,m;
Step S202, the index in original matrix X is pre-processed to obtain standard observation matrix Y;
Step S203, the coefficient R of standard observation matrix Y is solved;
Step S204, the characteristic root λ of correlation matrix R is solvedj, and characteristic root is sorted from big to small, determine principal component and master
The number p of component;
Step S205, the corresponding feature vector of principal component characteristic root and its principal component matrix F are extracted;
Step S206, determine in principal component matrix F, characteristic root λjCorresponding principal component column vector fjIt is described to the distance of ideal point
Ideal point is the maximum or minimum point of each row;
Step S207, calculated according to the distance to ideal point and be lined up indicated value, and trap queuing is carried out according to indicated value is lined up, obtained
To asset quality diagnostic result.
4. power grid enterprises' asset quality diagnosis according to claim 3 and method for early warning, it is characterised in that the pretreatment
Including:
1) to evaluation index xijUnification processing is carried out, is converted into large index;
2) nondimensionalization processing is carried out to the index for being converted into large, obtained by standard observation value yijThe standard observation square of composition
Battle array Y.
5. power grid enterprises' asset quality diagnosis according to claim 4 and method for early warning, it is characterised in that the unification
Processing includes two ways:
A. the index of minimal type, according to formula x 'ij=1-xij, it is converted into large index;
B. for the index of interval type, according to formulaLarge is converted into refer to
Mark;Wherein, [q1,q2] it is xijOptimal section, xminAnd xmaxFor xijThe minimum value and maximum of possible value.
6. power grid enterprises' asset quality diagnosis according to claim 4 and method for early warning, it is characterised in that the standard is seen
Measured valueIn formula,And sjThe respectively sample average and sample variance of jth item characteristic index value.
7. power grid enterprises' asset quality diagnosis according to claim 3 and method for early warning, it is characterised in that the principal component
Selection with λj> 1 and contribution rate of accumulative totalAs standard.
8. power grid enterprises' asset quality diagnosis according to claim 3 and method for early warning, it is characterised in that described to ideal
The distance of point includes fijTo the distance D of Positive ideal pointj +With the distance D to Negative ideal pointj -;
fijTo the distance of Positive ideal pointWherein, Positive ideal point zj +For column vector fjMaximum;
fijTo the distance of Negative ideal pointWherein;Negative ideal point zj -For column vector fjMinimum value;
Described calculate is lined up indicated value according to formulaObtain.
9. power grid enterprises' asset quality diagnosis according to claim 1 and method for early warning, it is characterised in that the step 3
Using random forest Early-warning Model as power grid enterprises' asset quality Early-warning Model.
10. power grid enterprises' asset quality diagnosis according to claim 9 and method for early warning, it is characterised in that
Step 3 includes following sub-step:
Step S301, the asset quality diagnostic result being calculated according to the principal component matrix F and principal component TOPSIS models
Classification, data set T of the structure new data set as random forest Early-warning Model;
Step S302, random forest Early-warning Model is trained using data set T;
Step S303, using the random forest Early-warning Model after training to the horizontal identification and classification of asset quality
Step S304, classified according to early warning, take different reply management measures.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711426676.7A CN108009751A (en) | 2017-12-26 | 2017-12-26 | A kind of power grid enterprises' asset quality diagnosis and method for early warning |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711426676.7A CN108009751A (en) | 2017-12-26 | 2017-12-26 | A kind of power grid enterprises' asset quality diagnosis and method for early warning |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108009751A true CN108009751A (en) | 2018-05-08 |
Family
ID=62061178
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711426676.7A Pending CN108009751A (en) | 2017-12-26 | 2017-12-26 | A kind of power grid enterprises' asset quality diagnosis and method for early warning |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108009751A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110110994A (en) * | 2019-05-06 | 2019-08-09 | 重庆大学 | Accurate trade and investment promotion business investment intention assessment system and method based on big data |
CN112529303A (en) * | 2020-12-15 | 2021-03-19 | 建信金融科技有限责任公司 | Risk prediction method, device, equipment and storage medium based on fuzzy decision |
-
2017
- 2017-12-26 CN CN201711426676.7A patent/CN108009751A/en active Pending
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110110994A (en) * | 2019-05-06 | 2019-08-09 | 重庆大学 | Accurate trade and investment promotion business investment intention assessment system and method based on big data |
CN112529303A (en) * | 2020-12-15 | 2021-03-19 | 建信金融科技有限责任公司 | Risk prediction method, device, equipment and storage medium based on fuzzy decision |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Chang et al. | Application of eXtreme gradient boosting trees in the construction of credit risk assessment models for financial institutions | |
Barboza et al. | Machine learning models and bankruptcy prediction | |
CN110956273A (en) | Credit scoring method and system integrating multiple machine learning models | |
CN107633265A (en) | For optimizing the data processing method and device of credit evaluation model | |
CN112417176B (en) | Method, equipment and medium for mining implicit association relation between enterprises based on graph characteristics | |
Hájek | Credit rating analysis using adaptive fuzzy rule-based systems: an industry-specific approach | |
CN108009751A (en) | A kind of power grid enterprises' asset quality diagnosis and method for early warning | |
Zhu et al. | Loan default prediction based on convolutional neural network and LightGBM | |
CN117522609A (en) | Bad financial recognition method, system and readable storage medium based on interpretable machine learning | |
Cui | Financial Credit Risk Warning Based on Big Data Analysis. | |
Kostopoulos et al. | Evaluating active learning methods for bankruptcy prediction | |
Wang et al. | Application of Data Mining Technology in Financial Data Analysis Methods Under the Background of Big Data | |
Yang et al. | An evidential reasoning rule-based ensemble learning approach for evaluating credit risks with customer heterogeneity | |
Zhang | Research on credit risk forecast model based on data mining technology | |
Guo et al. | Credit Default Prediction on Time-Series Behavioral Data Using Ensemble Models | |
Gu et al. | Financial Decision Management of Enterprise Cloud Accounting Based on Big Data Technology | |
Liu et al. | RETRACTED ARTICLE: Company financial path analysis using fuzzy c-means and its application in financial failure prediction | |
Guo et al. | Transductive Semi-Supervised Metric Network for Reject Inference in Credit Scoring | |
Said et al. | A Study on the Performance Comparison of Five Popular Machine Learning Models Applied for Loan Risk Prediction | |
Tan | Research on risk Prediction of enterprise Financial management based on optimized BP neural network algorithm | |
Yang | A Random Forest Approach to Appraise Personal Credit Risk of Internet Loans | |
Yang et al. | Research on Supply Chain Financial Risk Prevention Based on Machine Learning | |
Li et al. | Financial Intelligence Decision Analysis System Based on Decision Tree Algorithm | |
Ciobota et al. | Prediction of Business Bankruptcy with the Help of Extreme Gradient Increase. | |
Zhao | Research on Intelligent Analysis and Processing Technology of Financial Big Data Education Based on DM Algorithm |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180508 |