CN112037006A - Credit risk identification method and device for small and micro enterprises - Google Patents
Credit risk identification method and device for small and micro enterprises Download PDFInfo
- Publication number
- CN112037006A CN112037006A CN202010707012.3A CN202010707012A CN112037006A CN 112037006 A CN112037006 A CN 112037006A CN 202010707012 A CN202010707012 A CN 202010707012A CN 112037006 A CN112037006 A CN 112037006A
- Authority
- CN
- China
- Prior art keywords
- financial index
- financial
- enterprise
- classification
- value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 60
- 238000012549 training Methods 0.000 claims abstract description 60
- 230000002159 abnormal effect Effects 0.000 claims abstract description 46
- 238000012937 correction Methods 0.000 claims abstract description 6
- 238000005065 mining Methods 0.000 claims abstract description 4
- 238000001914 filtration Methods 0.000 claims description 25
- 238000007477 logistic regression Methods 0.000 claims description 15
- 238000012216 screening Methods 0.000 claims description 12
- 238000012163 sequencing technique Methods 0.000 claims description 8
- 238000012545 processing Methods 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 5
- 238000012986 modification Methods 0.000 claims description 5
- 230000004048 modification Effects 0.000 claims description 5
- 230000000717 retained effect Effects 0.000 claims description 5
- 238000013507 mapping Methods 0.000 claims description 4
- 238000010276 construction Methods 0.000 claims description 3
- 238000012360 testing method Methods 0.000 claims 2
- 230000009286 beneficial effect Effects 0.000 description 7
- 238000007726 management method Methods 0.000 description 6
- 238000007619 statistical method Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 5
- 238000004590 computer program Methods 0.000 description 4
- 238000012502 risk assessment Methods 0.000 description 4
- 230000007306 turnover Effects 0.000 description 4
- 230000007774 longterm Effects 0.000 description 3
- 230000015572 biosynthetic process Effects 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 238000003786 synthesis reaction Methods 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 238000012550 audit Methods 0.000 description 1
- 238000010224 classification analysis Methods 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 238000012417 linear regression Methods 0.000 description 1
- 230000007935 neutral effect Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000000611 regression analysis Methods 0.000 description 1
- 230000001502 supplementing effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/03—Credit; Loans; Processing thereof
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/12—Accounting
- G06Q40/125—Finance or payroll
Landscapes
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Engineering & Computer Science (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Technology Law (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The invention discloses a credit risk identification method and device for small and micro enterprises, which aim to improve the credit risk identification accuracy of the small and micro enterprises. The method comprises the following steps: mining sample data corresponding to the financial reports one by one based on the financial reports of a plurality of small micro-enterprises; data correction is carried out on the financial index classifications of null values and abnormal values in the sample data, and/or the financial index classifications are primarily screened and filtered according to the ratio of the number of the null values to the total amount of the sample data and the ratio of the number of the abnormal values to the total amount of the sample data in the same financial index classification; dividing each financial index classification into a plurality of sections according to the size of the corresponding financial index value, and matching the section of the financial index value in each sample data and the default rate of the corresponding enterprise to construct a training sample; training an enterprise default prediction model through the constructed training sample; and identifying the credit risk of the small and micro enterprise to be detected based on the financial report of the small and micro enterprise to be detected and the enterprise default prediction model.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to a credit risk identification method and device for small and micro enterprises.
Background
With the continuous development of internet technology, the technology of online credit for enterprises is also continuously perfected. The online development of credit business for enterprises requires the assessment of the credit of the enterprise, and there are many aspects of the factors for assessing the credit risk of the enterprise, including: financial information, business owner information, industrial and commercial information, pedestrian information, judicial complaint information and the like, wherein the risks possibly existing in the enterprises can be reflected to a certain extent in each aspect. Many scientific and technological enterprises in the market develop risk assessment models for enterprise credit assessment, and gradually form corresponding wind control theories. Although the financial information of the enterprises is applied to the risk assessment models in the training process, most of the risk assessment models only refer to the standard financial indexes compiled by the large and medium enterprises after supervision, that is, the non-standard financial indexes which can accurately reflect the risk conditions of the small and medium enterprises in the financial reports of the small and medium enterprises are not considered too much, so that the credit risk results of the small and medium enterprises identified by the risk assessment models are not accurate enough.
Disclosure of Invention
The invention aims to provide a credit risk identification method and device for small and micro enterprises, which combine a non-standard financial index training enterprise default prediction model capable of reflecting the risk condition of the small and micro enterprises with a standard financial index so as to improve the credit risk identification accuracy of the small and micro enterprises.
In order to achieve the above object, a first aspect of the present invention provides a credit risk identification method for a small micro-enterprise, including:
mining sample data corresponding to the financial reports one by one based on the financial reports of a plurality of small micro-enterprises, wherein the sample data comprises a plurality of financial index classifications of standard financial indexes and non-standard financial indexes;
data correction is carried out on the financial index classifications of null values and abnormal values in the sample data, and/or primary screening and filtering are carried out on the financial index classifications according to the ratio of the number of the null values to the total amount of the sample data and the ratio of the number of the abnormal values to the total amount of the sample data in the same financial index classification;
dividing each financial index classification into a plurality of sections according to the size of the corresponding financial index value, and matching the section of the financial index value in each sample data and the default rate of the corresponding enterprise to construct a training sample;
performing secondary filtration on the reserved financial index classifications according to the significance of each financial index classification in logistic regression analysis, and then training an enterprise default prediction model through the constructed training samples;
and identifying the credit risk of the small and micro enterprise to be detected based on the financial report of the small and micro enterprise to be detected and the enterprise default prediction model.
Illustratively, the standard financial indexes include a scale index, a profit index, a lever index, a liquidity index, an operation efficiency index, a liability coverage index and a growth index, and the non-standard financial indexes include an enterprise strength index, an operation management index, a liability capability index and an operation stability index.
Preferably, the method for data modification for the classification of the financial index which is an abnormal value in the sample data comprises:
extracting financial index values belonging to the same classification from the sample data respectively, and sequencing the financial index values corresponding to the financial index classifications in a descending order;
setting an abnormal value threshold value of each financial index classification, wherein the lower limit of the abnormal value threshold value is the A-th ordered financial index classification1The financial index values corresponding to the percentiles, and the upper limit of the abnormal value threshold is the A-th ordered financial index in the classification of the financial indexes2Financial index values corresponding to the percentiles;
replacing the financial index value with a lower limit of an abnormal value threshold if the financial index value in the sample data is lower than the lower limit of the abnormal value threshold of the belonging financial index classification, and replacing the financial index value with an upper limit of the abnormal value threshold if the financial index value in the sample data is higher than the upper limit of the abnormal value threshold of the belonging financial index classification.
Preferably, the method for data modification for the classification of the financial index which is null in the sample data includes:
extracting financial index values belonging to the same classification from the sample data respectively, and sequencing the financial index values corresponding to the financial index classifications in a descending order;
if the null value is caused by the loss of the financial reports, replacing the null value with the corresponding financial index scoreB-th ordered in class1Financial index values corresponding to the percentiles;
if the null value is generated due to variable calculation, replacing the null value with the B-th ordered in the financial index classification2And financial index values corresponding to the percentiles.
Preferably, the method for performing preliminary screening and filtering on the financial index classification according to the ratio of the number of null values to the total amount of sample data and the ratio of the number of abnormal values to the total amount of sample data in the same financial index classification includes:
when the ratio of the number of null values in the same classification to the total sample data exceeds a first threshold, primarily screening and filtering out the corresponding financial index classification;
and when the ratio of the number of the abnormal values to the total amount of the sample data of the same category exceeds a second threshold, primarily screening and filtering out the corresponding financial index categories.
Preferably, the method for classifying the financial indexes into a plurality of sections according to the sizes of the corresponding financial index values, and matching the sections of the financial index values in the sample data and the default rates of the corresponding enterprises to construct the training samples comprises:
acquiring a maximum financial index value and a minimum financial index value corresponding to each financial index classification in all sample data, and dividing the financial index value corresponding to each financial index classification into multiple sections;
and calculating corresponding enterprise default rates based on the sections where the financial index values corresponding to the financial index classifications are located, and constructing training samples according to the financial index classifications and the corresponding enterprise default rates.
Preferably, the method for training the enterprise default prediction model by the constructed training sample comprises the following steps of performing secondary filtering on the preserved financial index classifications according to the significance of each financial index classification in the logistic regression analysis:
calculating a KS value or an AUC value of each financial index classification pair training sample, and removing the financial index classifications below a KS threshold value or below an AUC threshold value;
calculating the significance value of the financial index classification in the logistic regression analysis method, and removing the financial index classification lower than the significance threshold value;
and training an enterprise default prediction model based on the training samples classified by the reserved financial indexes.
Preferably, the method for identifying the credit risk of the small-sized micro-enterprise to be tested based on the financial report of the small-sized micro-enterprise to be tested and the enterprise default prediction model comprises the following steps:
obtaining the preliminary prediction default probability P of the small and micro enterprise to be detected based on the financial newspaper of the small and micro enterprise to be detected and the enterprise default prediction model1；
Using the formula P-P2P1Calculating the predicted default probability of the small micro-enterprise to be tested, and matching the credit risk of the small micro-enterprise to be tested from a preset mapping relation based on the predicted default probability;
wherein, the P2To adjust the coefficients.
Preferably, after identifying the credit risk of the small-sized micro-enterprise to be tested based on the financial report of the small-sized micro-enterprise to be tested and the enterprise default prediction model, the method further comprises:
summarizing the filtered financial index classifications belonging to the non-standard financial index types in the sample data, and sequencing the financial index values corresponding to each financial index classification from small to large;
if the financial index value corresponding to any one of the financial index classifications is in a first interval percentile, the credit risk of the small micro-enterprise to be tested is adjusted down by one level until the credit risk is adjusted down to an access level;
if the financial index value corresponding to any one of the financial index classifications is lower than the lowest percentile, the credit risk of the small micro-enterprise to be tested is adjusted to be an admission level;
and if the financial index value corresponding to any one of the financial index classifications is in a second interval percentile, the credit risk of the small micro-enterprise to be tested is increased by one level until the credit risk is increased to the highest level.
Compared with the prior art, the credit risk identification method for the small micro-enterprise provided by the invention has the following beneficial effects:
according to the credit risk identification method for the small and micro enterprises, the adopted sample data not only comprise financial index classifications belonging to standard financial indexes, but also comprise financial index classifications corresponding to non-standard financial indexes capable of reflecting the risk characteristics of the small and micro enterprises, training samples are obtained after subsequent data correction and filtering screening, and finally, an enterprise default prediction model trained based on the training samples can accurately identify the credit risk of the small and micro enterprises.
A second aspect of the present invention provides a credit risk identification device for a small-sized micro-enterprise, which is applied to the above-mentioned credit risk identification method for a small-sized micro-enterprise, and the device includes:
the data acquisition unit is used for excavating sample data corresponding to the financial reports one by one on the basis of the financial reports of a plurality of small micro-enterprises, and the sample data comprises a plurality of financial index classifications of standard financial indexes and non-standard financial indexes;
the data processing unit is used for correcting data of the financial index classifications of null values and abnormal values in the sample data and/or primarily filtering the financial index classifications according to the ratio of the number of the null values to the total amount of the sample data and the ratio of the number of the abnormal values to the total amount of the sample data in the same financial index classification;
the sample construction unit is used for dividing each financial index classification into a plurality of sections according to the size of the corresponding financial index value, and matching the section where the financial index value is located in each sample data and the default rate of the corresponding enterprise to construct a training sample;
the model training unit is used for carrying out secondary filtering on the reserved financial index classifications according to the significance of each financial index classification in logistic regression analysis, and then training an enterprise default prediction model through the constructed training samples;
and the risk identification unit is used for identifying the credit risk of the small and micro enterprise to be detected based on the financial report of the small and micro enterprise to be detected and the enterprise default prediction model.
Compared with the prior art, the beneficial effects of the credit risk identification device for the small micro-enterprise provided by the invention are the same as the beneficial effects of the credit risk identification method for the small micro-enterprise provided by the technical scheme, and the details are not repeated herein.
A third aspect of the present invention provides a computer-readable storage medium, on which a computer program is stored, where the computer program is executed by a processor to execute the steps of the above-mentioned credit risk identification method for small enterprises.
Compared with the prior art, the beneficial effect of the computer-readable storage medium provided by the invention is the same as that of the credit risk identification method for small micro-enterprises provided by the technical scheme, and the detailed description is omitted here.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the invention and not to limit the invention. In the drawings:
FIG. 1 is a schematic flowchart illustrating a credit risk identification method for small and medium-sized enterprises according to an embodiment of the present invention;
FIG. 2 is a diagram illustrating an example of a relationship between a segment of financial index values and an enterprise default rate according to an embodiment of the present invention.
Detailed Description
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in detail below. It is to be understood that the described embodiments are merely exemplary of the invention, and not restrictive of the full scope of the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Example one
Referring to fig. 1, the present embodiment provides a credit risk identification method for a small enterprise, including:
mining sample data corresponding to the financial reports one by one based on the financial reports of a plurality of small micro-enterprises, wherein the sample data comprises a plurality of financial index classifications of standard financial indexes and non-standard financial indexes; data correction is carried out on the financial index classifications of null values and abnormal values in the sample data, and/or the financial index classifications are primarily screened and filtered according to the ratio of the number of the null values to the total amount of the sample data and the ratio of the number of the abnormal values to the total amount of the sample data in the same financial index classification; dividing each financial index classification into a plurality of sections according to the size of the corresponding financial index value, and matching the section of the financial index value in each sample data and the default rate of the corresponding enterprise to construct a training sample; performing secondary filtering on the reserved financial index classifications according to the significance of each financial index classification in logistic regression analysis, and then training an enterprise default prediction model through the constructed training samples; and identifying the credit risk of the small and micro enterprise to be detected based on the financial report of the small and micro enterprise to be detected and the enterprise default prediction model.
In the method for identifying credit risk of small and micro enterprises provided by this embodiment, the adopted sample data includes not only financial index classifications that belong to standard financial indexes, but also financial index classifications that correspond to non-standard financial indexes that can embody risk characteristics of small and micro enterprises, training samples are obtained after subsequent data correction and filtering, and finally, based on an enterprise default prediction model trained by these training samples, the credit risk of small and micro enterprises can be accurately identified.
The standard financial indexes comprise scale indexes, profit indexes, lever indexes, fluidity indexes, operation efficiency indexes, liability coverage indexes and growth indexes, and the non-standard financial indexes comprise enterprise strength indexes, operation management indexes, liability and liability ability indexes and operation stability indexes.
In specific implementation, the financial reports of small and medium enterprises need to be collected and sorted firstly, and different from large and medium enterprises, the financial reports of small and medium enterprises have the defects of nonstandard, various formats, deficiency and the like, so the collected financial reports need to be sorted according to a uniform format, generally, complete annual reports and latest quarterly financial reports in the last 2 years need to be collected, and if the financial reports are available, the financial reports can be pushed forward for one year.
The balance sheet in each small and micro enterprise financial newspaper is arranged according to the following format:
the profit and loss tables (profit tables) in each small and micro enterprise financial newspaper are arranged according to the following formats:
| item | Amount of money in this period | Amount of money in the past |
| Income of primary and main operation | ||
| Subtracting: cost of main business | ||
| Main business tax fund and additional | ||
| Profit of second, main business (loss filled with "-") | ||
| Adding: profits of other services (loss filled with "-") | ||
| Subtracting: business expenses | ||
| Managing fees | ||
| Financial cost | ||
| Third, business profit (loss filled with "-") | ||
| Adding: income from investment (loss filled with "-") | ||
| Income of subsidy | ||
| Income outside business | ||
| Fourthly, the total profit (the total loss is filled with a "-") | ||
| Subtracting: income tax fee | ||
| Five net profit (net loss filled with "-") |
And uniformly supplementing 0 to defect values and null values appearing in the table. And simultaneously checking the audit relationship of the two reports. The hooking and checking relationship comprises:
total assets + liabilities of owners;
total assets are liquidity + non-liquidity assets;
total liability is mobile liability + non-mobile liability;
the profit of the main business is the income of the main business, the cost of the main business, the tax and the addition of the main business;
the business profit is the profit of the main business plus the profit of other businesses, the business expense, the management expense and the financial expense;
the total profit is business profit, investment income, subsidy income and business income;
net profit-total profit-income tax fee, etc.
And then classifying and sorting the standard financial indexes, wherein the specific classifying and sorting process comprises the following steps: based on the common financial index classification mode in the market, classifying the financial indexes into the following 7 categories according to the meaning of the indexes and the embodied information, wherein each category corresponds to a specific financial index classification:
the financial indexes corresponding to the scale indexes are classified as follows: total assets, net assets, total sales, etc.;
the financial index classification corresponding to the profit index is as follows: total asset profitability, net profit, gross profit, capital profit, etc.;
the financial index classification corresponding to the lever index is as follows: asset liability rate, capitalization rate, net liability rate, short-term liability rate, long-term liability rate, etc.;
the financial indexes corresponding to the fluidity indexes are classified as follows: flow rate, snap rate, cash liability ratio, and the like;
the financial index classification that the operation efficiency class index corresponds to has: a mobile asset turnover rate, an accounts receivable turnover rate, an accounts payable turnover rate, a fixed asset turnover rate, etc.;
the financial index classification corresponding to the debt coverage index is as follows: EB. Cash debt ratio, gross income debt ratio, liquidity debt ratio, etc.;
the financial index categories corresponding to the growth index are as follows: total asset growth rate, net asset growth rate, sales growth rate, debt growth rate, etc.
The financial index classification analysis method is used for analyzing the financial index classifications of large and medium enterprises and then sorting out the corresponding financial index classifications and the corresponding financial index values based on the format of the financial statements. When the financial index values are sorted, the data are verified one by one according to the following sequence:
1. classifying the financial indexes which can be calculated by the subjects of the balance sheet and the profit-and-loss sheet at the same time, and averaging the subjects of the balance sheet for two years, such as total asset profitability (ROA);
2. when the financial index value calculated by dividing the two table subjects is found to be zero or negative, the denominator is set to be a larger value or a smaller value according to the economic significance of the index; if the denominator is zero or negative when the denominator is related to the subject of the debt (account payable, short term borrowing, long term borrowing, etc.), which indicates that the debt burden of the enterprise is small, the index can be set to a larger value, that is, the 99 th percentile value of the index of all samples is set as the value of the index after the index is sorted from small to large. If the denominator is the subject of one category of assets (such as flowing assets, cash, accounts receivable, net assets, etc.), the condition that the denominator is zero or negative indicates that the corresponding asset item of the enterprise has high small risk, and the index can be set to a smaller value, that is, the index of all samples is ranked from small to large and then the value of the 1 st percentile is set as the value of the index.
3. When a certain annual financial report is lacked, the corresponding growth rate index cannot be calculated, such as: the total asset growth rate, the net profit growth rate, the accounts receivable growth rate, the retained profit growth rate and the fixed asset growth rate are uniformly set to be replaced by corresponding index values according to 50 percent values after sorting from small to large.
Then, the non-standard financial indexes need to be classified and sorted, and the specific classification and sorting process is as follows: based on the common financial index classification mode in the market, classifying the general financial indexes into the following 4 categories according to the meaning of the indexes and the embodied information, wherein each category corresponds to a specific financial index classification:
the enterprise strength indexes comprise financial index classifications which reflect the sizes of enterprises and the investment conditions of enterprise owners, such as: actual capital income, retained income, tax intake and the like;
the operation management type indexes include indexes which can embody the operation management capacity and efficiency of enterprises in operation, such as the ratio of retained income to sales, the ratio of management cost to sales, the ratio of receivable accounts to sales, the ratio of payable accounts to sales, and the like;
the repayment capability index refers to the embodiment of the enterprise operation condition on the repayment capability of the existing debt, such as: the ratio of debts to sales, the ratio of debts to net assets, etc.;
the operation stability index refers to an index of the enterprise showing operation stability in actual operation, such as: stability of sales over the past twelve months, rate of increase of sales over the past half years, rate of increase of sales over the past three months, etc.;
the financial index classifications are not generally used when the large and medium enterprises are subjected to statistical analysis, or the large and medium enterprises have no obvious distinguishing capability, but the financial index classifications can reflect the operation condition and repayment capability of the small and medium enterprises, and are more suitable for being adopted by enterprise default prediction models of the small and medium enterprises.
In the above embodiment, the method for performing data modification for the financial index classification that is an abnormal value in sample data includes:
respectively extracting financial index values belonging to the same classification from each sample data, and sequencing the financial index values corresponding to each financial index classification from small to large;
setting abnormal value threshold of each financial index classification, wherein the lower limit of the abnormal value threshold is the A-th ordered financial index classification1The financial index values corresponding to the percentiles, and the upper limit of the abnormal value threshold is the A-th ordered in the classification of the financial indexes2Financial index values corresponding to the percentiles;
and if the financial index value in the sample data is lower than the lower limit of the abnormal value threshold of the belonged financial index classification, replacing the financial index value with the lower limit of the abnormal value threshold, and if the financial index value in the sample data is higher than the upper limit of the abnormal value threshold of the belonged financial index classification, replacing the financial index value with the upper limit of the abnormal value threshold.
In specific implementation, each sample data corresponds to multiple financial index classifications, the financial index classifications in each sample data are the same, and the financial index values of all sample data corresponding to each financial index classification are sorted from small to large; then, the abnormal value threshold corresponding to each financial index classification is set respectively, for example, A1Has a value of 1, A2The value of (1) is 99, that is, the financial index value lower than the 1 st percentile and the financial index value higher than the 99 th percentile can be treated as abnormal values, specifically, the financial index value lower than the 1 st percentile is replaced by the financial index value corresponding to the 1 st percentile, and the financial index value higher than the 99 th percentile is replaced by the financial index value corresponding to the 99 th percentile.
In the above embodiment, the method for correcting data for the financial index classification that is null in the sample data includes:
respectively extracting financial index values belonging to the same classification from each sample data, and sequencing the financial index values corresponding to each financial index classification from small to large; if the null value is caused by the missing financial index, replacing the null value with the B-th ordered financial index in the classification of the financial index1Financial index values corresponding to the percentiles; if the null value is generated due to variable calculation, replacing the null value with the B-th ordered in the financial index classification2And financial index values corresponding to the percentiles.
In practice, if the null value of the financial index is due to the absence of normal data, such absence can be directly attributed to a neutral value; for example, if the annual financial report is lacked, various growth rate indexes, such as total asset growth rate, net profit growth rate, receivable account growth rate, retained profit growth rate, fixed asset growth rate and the like, cannot be calculated, the lacked financial index values are replaced by the financial index values corresponding to the 50 th percentile, namely the financial index values B1Is 50.
If the financial index value isThe null value is caused by abnormal calculation during variable synthesis, if the denominator is zero or negative, the null value is processed according to the situation, if the denominator is debt, such as account payable, short term borrowing, long term borrowing and other related subjects, the situation that the denominator is zero or negative indicates that the debt burden of the enterprise is small, and then the missing financial index values can be replaced by the financial index value corresponding to the 99 th percentile, namely B2Is 99. If the denominator is an asset, such as a mobile asset, cash, accounts receivable, clean asset, and other subjects, the condition that the denominator is zero or negative indicates that the corresponding asset item of the enterprise is low-risk, and then the financial index values can be replaced by the financial index values corresponding to the 1 st percentile, that is, the financial index values B2Is 1.
In the above embodiment, the method for performing preliminary screening and filtering on the financial index classification according to the ratio of the number of null values to the total amount of sample data and the ratio of the number of abnormal values to the total amount of sample data in the same financial index classification includes:
when the ratio of the number of null values in the same classification to the total sample data exceeds a first threshold, primarily screening and filtering out the corresponding financial index classification; and when the ratio of the number of the abnormal values to the total amount of the sample data of the same category exceeds a second threshold, primarily screening and filtering out the corresponding financial index categories.
In specific implementation, if the abnormal value corresponding to a certain financial index classification is too much, for example, more than 50% of the total amount of all sample data, the financial index classification is discarded. If the null value corresponding to a certain financial index classification is too much, such as more than 80% of the total amount of all sample data, the financial index classification is rejected. That is, the value of the first threshold is 50%, and the value of the second threshold is 80%.
In the above embodiment, the method for classifying the financial indexes into multiple segments according to the sizes of the corresponding financial index values, and matching the segments where the financial index values are located in each sample data and the default rate of the corresponding enterprise to construct the training samples includes:
acquiring a maximum financial index value and a minimum financial index value corresponding to each financial index classification in all sample data, and dividing the financial index values corresponding to each financial index classification into multiple sections; and calculating corresponding enterprise default rates based on the sections where the financial index values corresponding to the financial index classifications are located, and constructing training samples according to the financial index classifications and the corresponding enterprise default rates.
In specific implementation, each financial index classification in sample data needs to be subjected to data processing respectively before regression analysis is performed, the specific process is that a maximum financial index value and a minimum financial index value corresponding to each financial index classification are obtained from all sample data, then the difference value between the maximum financial index value and the minimum financial index value of each financial index classification is calculated, if the difference value is divided into 10 equal parts, the financial index value corresponding to each financial index classification is finally divided into 10 sections, then the corresponding enterprise default rate is calculated based on the financial index classification to which each financial index value belongs and the section where each financial index value belongs, and a training sample is constructed until the enterprise default rate corresponding to each financial index value in all sample data is calculated. As shown in fig. 2, the enterprise default rate of the segment 1-2 to which the financial index classification a belongs is 7.5%, the enterprise default rate of the segment 2-3 to which the financial index classification a belongs is 6%, and finally, a training sample is constructed according to each financial index classification and the corresponding enterprise default rate.
In the above embodiment, the method for training the enterprise default prediction model by using the constructed training samples after performing secondary filtering on the remaining financial index classifications according to the significance of each financial index classification in the logistic regression analysis includes:
calculating a KS value or an AUC value of each financial index classification pair training sample, and removing the financial index classifications below a KS threshold value or below an AUC threshold value; calculating the significance value of the financial index classification in the logistic regression analysis method, and removing the financial index classification lower than the significance threshold value; and training an enterprise default prediction model based on the training samples classified by the reserved financial indexes.
In the specific implementation, the financial index classifications are preliminarily screened through a non-parameter statistic KS value or AUC value, the KS value or AUC value of a training sample is calculated for each financial index classification, if the KS value or AUC value is lower than a KS threshold value or an AUC threshold value, the KS threshold value is 0.08, and the AUC threshold value is 0.55, the financial index classifications are removed, in the logistic regression analysis process, the financial index classifications with the significance p value lower than the significance threshold value are removed by a conventional method, the significance threshold value is 3, so that the data in the training sample are updated, and then the enterprise default prediction model is trained through a linear regression method or a logistic regression method based on the updated training sample.
It can be understood that the training method of the enterprise default prediction model is the prior art in the field, and details thereof are not described in this embodiment.
In addition, the enterprise default prediction model can be verified by the following method: the relevance of the financial index classification is checked, i.e. a variable collinearity check. If the result is checked by a VIF (variance Information factor) variance expansion coefficient, the VIF can directly obtain a result through statistical software during logistic regression, if the calculation result of the VIF is more than 5, the financial index classification needs to be readjusted, and otherwise, the adjustment is not needed. And under the condition that the financial index classification in the training data is not required to be adjusted, aiming at the filtered financial index classification belonging to the non-standard financial index type, calculating a KS value or an AUC value of the financial index classification, if the KS value or the AUC value is reasonable, namely the KS value or the AUC value is larger than a preset KS threshold or AUC threshold, recognizing the model, and otherwise, adjusting or redeveloping the model.
In the above embodiment, the method for identifying the credit risk of the small-sized micro-enterprise to be tested based on the financial report of the small-sized micro-enterprise to be tested and the enterprise default prediction model includes:
obtaining the preliminary forecast default probability P of the small and micro enterprise to be tested based on the financial newspaper of the small and micro enterprise to be tested and the enterprise default forecast model1(ii) a Using the formula P ═ P2*P1Calculating the predicted default probability of the small micro-enterprise to be tested, and matching the credit risk of the small micro-enterprise to be tested from the preset mapping relation based on the predicted default probability; wherein, P2To adjust the coefficients.
In specific implementation, sample data with the same dimension as the updated training sample is obtained from the financial newspaper of the small and micro enterprise to be tested, and the sample data is updatedInputting the sample data into a trained enterprise default prediction model to obtain a preliminary prediction default probability P1Wherein, in the step (A),
the P isPractice ofRepresenting the mean value of the default rate of the corresponding enterprise of each segment occupied by the financial index values in the sample data, wherein P isSample(s)Representing the ratio of default enterprises to the total amount of training samples. Because a mapping relation table of the predicted default probability and the credit risk is preset, the credit risk of the small micro-enterprise to be tested can be automatically matched and output according to the finally obtained predicted default probability.
In the above embodiment, after identifying the credit risk of the small-sized micro-enterprise to be tested based on the financial report of the small-sized micro-enterprise to be tested and the enterprise default prediction model, the method further includes:
the filtered financial index classifications which belong to the non-standard financial index types in the sample data are summarized, and the financial index values corresponding to each financial index classification are sequenced from small to large; if the financial index value corresponding to any financial index classification is in the first interval percentile, the credit risk of the small micro-enterprise to be tested is reduced by one level, if the financial index values corresponding to a plurality of financial index classifications are in the first interval percentile, the credit risk is correspondingly reduced by a plurality of levels until the credit risk is reduced to the access level; if the financial index value corresponding to any financial index classification is lower than the lowest percentile, the credit risk of the small and micro enterprise to be tested is adjusted to be a permitted entry level; and if the financial index values corresponding to any financial index classification are in the second interval percentile, the credit risk of the small micro-enterprise to be tested is increased by one level, and if the financial index values corresponding to a plurality of financial index classifications are in the second interval percentile, the credit risk is correspondingly increased by a plurality of levels until the credit risk is increased to the highest level.
In specific implementation, if the financial index value corresponding to any financial index classification is in a first interval percentile, the credit risk of the small micro-enterprise to be tested is reduced by one level until the credit risk is reduced to an admission level, wherein the first interval is an interval of 2 to 15 percentiles, such as AA-level and A + level; if the financial index value corresponding to any financial index classification is lower than the lowest percentile, the credit risk of the small and micro enterprise to be tested is adjusted to be a permitted entry level, and the lowest percentile is 2 percentile; if the financial index value corresponding to any financial index classification is in the second interval percentile, the credit risk of the small micro-enterprise to be tested is increased by one level until the credit risk is increased to the highest level, and the first interval is 15-30 percentile; and (4) adjusting the classification missing value of the financial index, and if the financial statement of the previous year is missing, the credit risk of the enterprise is adjusted down by one level.
In summary, in the prior art, when developing a risk model for a small micro enterprise, the following data sources are mainly considered: financial statement information, tax, people's bank credit, social bank credit, industry and commerce judicial information, etc. However, the design and synthesis of related variables of the financial index classification are not obviously different from large and medium enterprises, and the characteristics of small and medium enterprises are not highlighted. Because the financial statements of the small and micro enterprises are incomplete, the information is unreliable, and the conventional standard financial indexes are considered in the incomplete places, the non-standard financial indexes suitable for the small and micro enterprises need to be considered and designed in a targeted manner. Furthermore, when the conventional statistical method selects the financial index classification with risk differentiation capability, the statistical universal indexes are more biased, and some risk indexes which are low in occurrence probability but are fatal or very critical to the credit of the small micro-enterprise are excluded. Or the indexes are not obvious in statistical analysis and are easy to screen by a statistical method.
The embodiment aims to break out of a frame of conventional risk factors, design some financial index classifications which can better reflect the risk of the small and micro enterprise for statistical analysis based on the fact that the reliability of data is low, then find out important risk factors with low statistical significance based on the statistical analysis result, and enter a model through different adjustment modes. The model developed in this way can predict credit risk of small micro-enterprises more accurately.
Example two
The embodiment provides a credit risk recognition device of small and micro enterprise, including:
the data acquisition unit is used for excavating sample data corresponding to the financial reports one by one on the basis of the financial reports of a plurality of small micro-enterprises, and the sample data comprises a plurality of financial index classifications of standard financial indexes and non-standard financial indexes;
the data processing unit is used for correcting data of the financial index classifications of null values and abnormal values in the sample data and/or primarily filtering the financial index classifications according to the ratio of the number of the null values to the total amount of the sample data and the ratio of the number of the abnormal values to the total amount of the sample data in the same financial index classification;
the sample construction unit is used for dividing each financial index classification into a plurality of sections according to the size of the corresponding financial index value, and matching the section where the financial index value is located in each sample data and the default rate of the corresponding enterprise to construct a training sample;
the model training unit is used for carrying out secondary filtering on the reserved financial index classifications according to the significance of each financial index classification in logistic regression analysis, and then training an enterprise default prediction model through the constructed training samples;
and the risk identification unit is used for identifying the credit risk of the small and micro enterprise to be detected based on the financial report of the small and micro enterprise to be detected and the enterprise default prediction model.
Compared with the prior art, the beneficial effects of the credit risk identification device for the small-sized micro-enterprise provided by the embodiment of the invention are the same as the beneficial effects of the credit risk identification method for the small-sized micro-enterprise provided by the embodiment one, and are not repeated herein.
EXAMPLE III
The embodiment provides a computer-readable storage medium, on which a computer program is stored, and the computer program is executed by a processor to perform the steps of the above-mentioned credit risk identification method for small enterprises.
Compared with the prior art, the beneficial effect of the computer-readable storage medium provided by the embodiment is the same as that of the credit risk identification method for small enterprises provided by the technical scheme, and details are not repeated herein.
It will be understood by those skilled in the art that all or part of the steps in the method for implementing the invention may be implemented by hardware instructions related to a program, the program may be stored in a computer-readable storage medium, and when executed, the program includes the steps of the method of the embodiment, and the storage medium may be: ROM/RAM, magnetic disks, optical disks, memory cards, and the like.
The above description is only for the specific embodiments of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present invention, and the changes or substitutions should be covered within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the appended claims.
Claims (10)
1. A credit risk identification method for small micro-enterprises is characterized by comprising the following steps:
mining sample data corresponding to the financial reports one by one based on the financial reports of a plurality of small micro-enterprises, wherein the sample data comprises a plurality of financial index classifications of standard financial indexes and non-standard financial indexes;
performing data processing on the sample to perform primary screening and filtering on the financial index classification;
dividing each financial index classification into a plurality of sections according to the size of the corresponding financial index value, and matching the section of the financial index value in each sample data and the default rate of the corresponding enterprise to construct a training sample;
performing secondary filtration on the reserved financial index classifications according to the significance of each financial index classification in logistic regression analysis, and then training an enterprise default prediction model through the constructed training samples;
and identifying the credit risk of the small and micro enterprise to be detected based on the financial report of the small and micro enterprise to be detected and the enterprise default prediction model.
2. The method of claim 1, wherein the step of performing data processing on the sample to perform preliminary filtering of the financial index classifications comprises:
and performing data correction on the financial index classifications of null values and abnormal values in the sample data, and/or performing primary screening and filtering on the financial index classifications according to the ratio of the number of the null values to the total amount of the sample data and the ratio of the number of the abnormal values to the total amount of the sample data in the same financial index classification.
3. The method of claim 2, wherein the method of data modification for a classification of financial indicators in the sample data that are outliers comprises:
extracting financial index values belonging to the same classification from the sample data respectively, and sequencing the financial index values corresponding to the financial index classifications in a descending order;
setting an abnormal value threshold value of each financial index classification, wherein the lower limit of the abnormal value threshold value is the A-th ordered financial index classification1The financial index values corresponding to the percentiles, and the upper limit of the abnormal value threshold is the A-th ordered financial index in the classification of the financial indexes2Financial index values corresponding to the percentiles;
replacing the financial index value with a lower limit of an abnormal value threshold if the financial index value in the sample data is lower than the lower limit of the abnormal value threshold of the belonging financial index classification, and replacing the financial index value with an upper limit of the abnormal value threshold if the financial index value in the sample data is higher than the upper limit of the abnormal value threshold of the belonging financial index classification.
4. The method of claim 2, wherein the method of data modification for the classification of financial indicators in the sample data that are null comprises:
extracting financial index values belonging to the same classification from the sample data respectively, and sequencing the financial index values corresponding to the financial index classifications in a descending order;
if the null value is caused by the missing financial reportAnd if so, replacing the null value with the B-th ordered in the financial index classification1Financial index values corresponding to the percentiles;
if the null value is generated due to variable calculation, replacing the null value with the B-th ordered in the financial index classification2And financial index values corresponding to the percentiles.
5. The method of claim 1, wherein the step of prescreening the financial index classifications based on a ratio of a number of null values to a total amount of sample data and a ratio of a number of outlier values to a total amount of sample data for the same financial index classification comprises:
when the ratio of the number of null values in the same classification to the total sample data exceeds a first threshold, primarily screening and filtering out the corresponding financial index classification;
and when the ratio of the number of the abnormal values to the total amount of the sample data of the same category exceeds a second threshold, primarily screening and filtering out the corresponding financial index categories.
6. The method of claim 1, wherein the step of classifying the financial index into a plurality of segments according to the size of the corresponding financial index, and matching the segment of the financial index in each sample data and the corresponding default rate of the enterprise to construct the training sample comprises:
acquiring a maximum financial index value and a minimum financial index value corresponding to each financial index classification in all sample data, and dividing the financial index value corresponding to each financial index classification into multiple sections;
and calculating corresponding enterprise default rates based on the sections where the financial index values corresponding to the financial index classifications are located, and constructing training samples according to the financial index classifications and the corresponding enterprise default rates.
7. The method of claim 1, wherein the retained financial index classifications are secondarily filtered according to the significance of each of the financial index classifications in the logistic regression analysis, and the method of training the business default prediction model using the constructed training samples comprises:
calculating a KS value or an AUC value of each financial index classification pair training sample, and removing the financial index classifications below a KS threshold value or below an AUC threshold value;
calculating the significance value of the financial index classification in the logistic regression analysis method, and removing the financial index classification lower than the significance threshold value;
and training an enterprise default prediction model based on the training samples classified by the reserved financial indexes.
8. The method of claim 1, wherein the method for identifying the credit risk of the small-sized micro-enterprise to be tested based on the financial report of the small-sized micro-enterprise to be tested and the enterprise default prediction model comprises:
obtaining the preliminary prediction default probability P of the small and micro enterprise to be detected based on the financial newspaper of the small and micro enterprise to be detected and the enterprise default prediction model1；
Using the formula P ═ P2*P1Calculating the predicted default probability of the small micro-enterprise to be tested, and matching the credit risk of the small micro-enterprise to be tested from a preset mapping relation based on the predicted default probability;
wherein, the P2To adjust the coefficients.
9. The method of claim 1, wherein identifying the credit risk of the small-sized micro-enterprise under test based on the financial report of the small-sized micro-enterprise under test and the enterprise default prediction model further comprises:
summarizing the filtered financial index classifications belonging to the non-standard financial index types in the sample data, and sequencing the financial index values corresponding to each financial index classification from small to large;
if the financial index value corresponding to any one of the financial index classifications is in a first interval percentile, the credit risk of the small micro-enterprise to be tested is adjusted down by one level until the credit risk is adjusted down to an access level;
if the financial index value corresponding to any one of the financial index classifications is lower than the lowest percentile, the credit risk of the small micro-enterprise to be tested is adjusted to be an admission level;
and if the financial index value corresponding to any one of the financial index classifications is in a second interval percentile, the credit risk of the small micro-enterprise to be tested is increased by one level until the credit risk is increased to the highest level.
10. A credit risk identification device for small micro-enterprises, comprising:
the data acquisition unit is used for excavating sample data corresponding to the financial reports one by one on the basis of the financial reports of a plurality of small micro-enterprises, and the sample data comprises a plurality of financial index classifications of standard financial indexes and non-standard financial indexes;
the data processing unit is used for correcting data of the financial index classifications of null values and abnormal values in the sample data and/or primarily filtering the financial index classifications according to the ratio of the number of the null values to the total amount of the sample data and the ratio of the number of the abnormal values to the total amount of the sample data in the same financial index classification;
the sample construction unit is used for dividing each financial index classification into a plurality of sections according to the size of the corresponding financial index value, and matching the section where the financial index value is located in each sample data and the default rate of the corresponding enterprise to construct a training sample;
the model training unit is used for carrying out secondary filtering on the reserved financial index classifications according to the significance of each financial index classification in logistic regression analysis, and then training an enterprise default prediction model through the constructed training samples;
and the risk identification unit is used for identifying the credit risk of the small and micro enterprise to be detected based on the financial report of the small and micro enterprise to be detected and the enterprise default prediction model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010707012.3A CN112037006A (en) | 2020-07-21 | 2020-07-21 | Credit risk identification method and device for small and micro enterprises |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010707012.3A CN112037006A (en) | 2020-07-21 | 2020-07-21 | Credit risk identification method and device for small and micro enterprises |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112037006A true CN112037006A (en) | 2020-12-04 |
Family
ID=73579345
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010707012.3A Pending CN112037006A (en) | 2020-07-21 | 2020-07-21 | Credit risk identification method and device for small and micro enterprises |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112037006A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112767103A (en) * | 2020-12-30 | 2021-05-07 | 北京知因智慧科技有限公司 | Financial data analysis method and device and electronic equipment |
| CN113780582A (en) * | 2021-09-15 | 2021-12-10 | 杭银消费金融股份有限公司 | Wind control feature screening method and system based on machine learning model |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110246031A (en) * | 2019-06-21 | 2019-09-17 | 深圳前海微众银行股份有限公司 | Appraisal procedure, system, equipment and the storage medium of business standing |
| CN111192140A (en) * | 2020-01-02 | 2020-05-22 | 北京明略软件系统有限公司 | Method and device for predicting customer default probability |
-
2020
- 2020-07-21 CN CN202010707012.3A patent/CN112037006A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110246031A (en) * | 2019-06-21 | 2019-09-17 | 深圳前海微众银行股份有限公司 | Appraisal procedure, system, equipment and the storage medium of business standing |
| CN111192140A (en) * | 2020-01-02 | 2020-05-22 | 北京明略软件系统有限公司 | Method and device for predicting customer default probability |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112767103A (en) * | 2020-12-30 | 2021-05-07 | 北京知因智慧科技有限公司 | Financial data analysis method and device and electronic equipment |
| CN113780582A (en) * | 2021-09-15 | 2021-12-10 | 杭银消费金融股份有限公司 | Wind control feature screening method and system based on machine learning model |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2012018968A1 (en) | Method and system for quantifying and rating default risk of business enterprises | |
| CN101194286A (en) | Risk based data assessment | |
| CN112598500A (en) | Credit processing method and system for non-limit client | |
| CN111476660A (en) | Intelligent wind control system and method based on data analysis | |
| CN110909963A (en) | Credit scoring card model training method and taxpayer abnormal risk assessment method | |
| CN112102073A (en) | Credit risk control method and system, electronic device and readable storage medium | |
| CN111401798A (en) | Enterprise waste escaping and debt risk early warning system and construction method | |
| CN110689437A (en) | Communication construction project financial risk prediction method based on random forest | |
| CN112613977A (en) | Personal credit loan admission credit granting method and system based on government affair data | |
| CN113095927A (en) | Method and device for identifying suspicious transactions of anti-money laundering | |
| CN112037006A (en) | Credit risk identification method and device for small and micro enterprises | |
| CN111090833A (en) | Data processing method, system and related equipment | |
| CN111626855A (en) | Bond credit interest difference prediction method and system | |
| CN116596674A (en) | External trade risk assessment method based on big data analysis | |
| CN115860924A (en) | Supply chain financial credit risk early warning method and related equipment | |
| CN108197740A (en) | Business failure Forecasting Methodology, electronic equipment and computer storage media | |
| CN113298575A (en) | Method, system, equipment and storage medium for trademark value batch evaluation | |
| CN111612602A (en) | Suspected financial risk distinguishing method and device for listed company | |
| CN112767121A (en) | Method and device for processing risk level data | |
| CN112685563A (en) | Cash flow data processing method and device under production and operation view | |
| CN111951105A (en) | Intelligent credit wind control system based on multidimensional big data analysis | |
| Andersson et al. | Bankruptcy determinants among Swedish SMEs:-The predictive power of financial measures | |
| Anissa et al. | ANALYSIS OF IMPLEMENTATION OF PSAK 71 AGAINST RESERVE LOSS OF IMPAIRMENT WHEN COVID-19 PANDEMIC IN PT. BANK CENTRAL ASIA TBK | |
| Dedu et al. | Developing a rating model on a statistical basis | |
| CN118071482A (en) | Method for constructing retail credit risk prediction model and consumer credit business Scorebetad model |
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: 20201204 |