TWI658423B - Dynamically adjustable loan system and method - Google Patents

Abstract

A loan system includes a credit rating detection module, a repayment process detection module, a reserve water level detection module, a risk allocation ratio control module, a default payment ratio control module, and a reserve water level control server . The reserve water level control server has a database. The reserve water level detection module is used to calculate the estimated total compensation and reserve difference based on the remaining outstanding principal and default rate of each of all the lenders in the database. The risk allocation ratio control module is used to determine the risk allocation ratio of the newly added lender according to the default rate and the reserve difference of the new lender. The reserve water level control server sends instructions to the repayment process on a regular basis. Detection module and reserve water level detection module in order to update the reserve balance, and then dynamically adjust the risk allocation ratio and the default compensation ratio.

Description

Dynamically adjustable lending system and method

The invention relates to the field of e-commerce, and in particular to a loan system and method.

In the existing lending structure, the borrower has a certain default rate based on his own credit rating. Since the investor does not want to lose the investment, he requires the borrower to allocate a certain amount of money in accordance with the estimated repayment risk In order to compensate investors when a default occurs.

In response, existing loan systems use traditional insurance methods to calculate static default rates. Specifically, the traditional insurance method uses historical statistical data to calculate static optimization parameters. However, due to the rapid spread of the Internet, there is a large amount of trading volume every day, and static optimization parameters cannot be used to immediately reflect changes in the environment, fail to maintain default compensation operations, and cause poor risk management of the lending system.

In view of this, the present invention provides a loan system, including Including: a reserve water level control server with a database, the reserve water level control server is used to receive a loan application from a lender's device, wherein the loan application includes information about the newly added lender; credit rating The detection module is connected to the reserve water level control server, and is configured to receive the data of the newly added lender from the reserve water level control server, and according to the data of the newly added lender Determine the credit rating and default rate of the newly added lender, in order to store the credit rating and default rate of the new lender, in the database; the repayment process detection module, connect The reserve water level control server is used for receiving instructions from the reserve water level control server to detect the repayment status of each of all the lenders in the database and calculate each of the lenders. And the remaining unpaid principal is stored in the database; a reserve water level detection module is connected to the reserve water level control server for receiving the reserve water level control servo Finger And calculate the estimated total compensation and the reserve difference based on the remaining outstanding principal and the default rate of each of all the lenders in the database, and the estimated total compensation and the reserve difference Stored in the database; the risk allocation ratio control module is connected to the reserve water level control server and is used to receive instructions from the reserve water level control server and according to the newly applied loan in the database The default rate and the reserve difference of the person determine the risk provision ratio of the new lender, and store the risk provision ratio of the new lender in the database; and the default payment The proportional control module is connected to the reserve water level control server, and is used for receiving instructions from the reserve water level control server and adjusting the office in the database according to the reserve balance of the database. There is a default payment ratio for each of the loan applicants; among them, the reserve water level control server periodically sends instructions to the repayment process detection module and the reserve water level detection module in order to update the data in the database. The reserve balance further dynamically adjusts the risk allocation ratio and the default compensation ratio in the database.

The present invention also provides a method for lending by computer, including the following steps: a reserve water level control server receives a loan application from a lender's device, wherein the loan application includes information of the newly added lender; credit evaluation The detection module receives the data of the newly added lender from the reserve water level control server, and determines the credit rating of the newly added lender according to the information of the new lender. And the default rate, and the credit rating of the newly added lender and the default rate are stored in the database of the reserve water level control server; the risk allocation ratio control module receives the reserve water level Control the server's instruction to determine the risk allocation ratio of the new lender based on the default rate and reserve difference of the new lender, and to increase the risk of the new lender The allocation ratio is stored in the database; and the default payout ratio control module receives an instruction from the reserve water level control server and determines the newly-added loan application based on the credit rating of the newly-added loan applicant. And the default payout ratio of the newly added lender is stored in the database; wherein the reserve difference is based on the remaining outstanding principal of each of all lenders, the The default rate and reserve account balance of each of all lenders are obtained; and among them, the risk provision ratio of the newly added lender and the all of them are dynamically adjusted by periodically calculating the reserve difference. Shen The lender's default compensation ratio.

The present invention further provides a computer-assisted lending method including the following steps: The repayment process detection module receives instructions from the reserve water level control server and periodically checks all applications in the database of the reserve water level control server. The repayment status of each of the lenders determines whether there is a default; when one of the all lenders defaults, the reserve water level detection module pays based on the default lender's default compensation ratio in the database. The corresponding investor; after paying the investor, the reserve water level detection module receives an instruction from the reserve water level control server and according to the remaining outstanding principal of each of all the lenders, Calculate the reserve balance by the default rate and the reserve account balance of each of the remaining lenders; of which, the risk provision of the newly added lenders is dynamically adjusted by periodically calculating the reserve difference Ratio and the default payment ratio of all the lenders.

The invention provides an online platform that allows loan applicants and investors to freely match lending transactions. By dynamically adjusting the risk allocation ratio and the default compensation ratio, the newly added application is immediately allowed to maintain the reserve level. The lender knows the cost that it should pay, and the investor can also obtain dynamically adjusted compensation when the lender of credit defaults, thereby maintaining the default compensation operation. The invention also provides a dynamically adjustable lending system and method that can be adjusted automatically by a computer, so that it can respond to changes in the environment in real time, so that the lending system is dynamically optimized, and the stable operation of the lending system is maintained. .

100‧‧‧Lending system

110‧‧‧Credit rating detection module

120‧‧‧ Repayment process detection module

130‧‧‧ Reserve water level detection module

140‧‧‧Risk allocation ratio control module

150‧‧‧ Default Payment Ratio Control Module

160‧‧‧ Reserve Water Level Control Server

162‧‧‧Database

170‧‧‧ Reserve Account

180‧‧‧Economic boom detection module

190‧‧‧Community Behavior Detection Module

S202 ~ S214‧‧‧step

S302 ~ S316‧‧‧step

S402 ~ S406‧‧‧step

S502 ~ S516‧‧‧step

S602 ~ S614‧‧‧step

Figure 1 is a system schematic diagram of the disclosed lending system; Figure 2 is a credit rating detection flowchart of the disclosed lending system; Figure 3 is a repayment process detection flowchart of the disclosed lending system; Figure 4 is The flow chart of the reserve water level detection of the disclosed lending system; Figure 5 is a flowchart of the newly added lenders of the disclosed lending system; and Figure 6 is the control flowchart of the default payment ratio of the disclosed lending system.

FIG. 1 is a system diagram of the disclosed loan system. The loan system 100 includes a credit rating detection module 110, a repayment process detection module 120, a reserve water level detection module 130, a risk allocation ratio control module 140, a default payment ratio control module 150, and a reserve Water level control server 160 and reserve account 170. The reserve water level control server 160 has a database 162, and is connected to the credit rating detection module 110, the repayment process detection module 120, the reserve water level detection module 130, the risk allocation ratio control module 140, and Default compensation ratio control module 150. The reserve water level detection module 130 is connected to the banking system so as to obtain the balance of the reserve account 170 in the banking system. In some embodiments, the loan system 100 further includes an economic prosperity detection module 180 and a community behavior detection module 190. The economic prosperity detection module 180 and the social behavior detection module 190 are respectively connected to the reserve water level control. Server 160.

In detail, the lending system 100 may be any electronic device with calculation and storage functions, such as a notebook computer, a desktop computer, a server, and the like. In other embodiments, the loan system 100 may also be composed of multiple electronic devices. In addition, in some embodiments, the credit rating detection module 110, the repayment process detection module 120, the reserve water level detection module 130, the risk allocation ratio control module 140, and the default payment ratio control module 150. The reserve water level control server 160, the economic prosperity detection module 180, and the community behavior detection module 190 are independent electronic devices. In other embodiments, the credit rating detection module 110, Repayment process detection module 120, reserve water level detection module 130, risk allocation ratio control module 140, default compensation ratio control module 150, reserve water level control server 160, economic boom detection module 180 And the community behavior detection module 190 is integrated into the same device. In some embodiments, the lending system 100 is a lending platform applied to the Internet. Lenders and investors use the network to mediate lending transactions on this lending platform.

The credit rating detection module 110 is configured to receive data of the new creditor from the reserve water level control server 160, and determine the credit rating of the new creditor based on the information of the new creditor. And default rate in order to store the credit rating and default rate of the newly added lenders in the database 162. In some embodiments, the information of the newly added lenders includes at least one of the newly added lenders' credit report records, repayment records, total loan amount, personal income, work company and personal expenses.

Specifically, please refer to FIG. 2 at the same time. FIG. 2 is a credit rating detection flow chart of the loan system 100 disclosed. Lending system 100 The reserve water level control server 160 receives a loan application from the newly-added loan applicant n + 1, wherein the loan application includes data of the loan applicant n + 1, and then the credit rating detection module 110 determines the loan applicant n + 1 Credit rating e (n + 1) and default rate c (n + 1). As shown in FIG. 2, in step S202, an initial value of the credit rating e (n + 1) is set according to the credit report record of the creditor n + 1, where the initial value is a preset intermediate number. In addition, the credit report record can be obtained by applying for credit to the credit reporting agency n + 1. In one embodiment, a credit check is performed on the lender n + 1 according to the credit report record of the lender n + 1. When the lender n + 1 fails the credit check, the lender n + 1 is rejected. loan application. In step S204, the repayment record of the credit lender n + 1 is checked, and the credit rating e (n + 1) is adjusted according to the situation. Among them, if the credit lender n + 1 has a record of late payment of loans or bills, it is reduced. Credit rating e (n + 1). In step S206, the total loan amount of the credit lender n + 1 is checked, and the credit rating e (n + 1) is adjusted according to the situation. Among them, the credit rating is adjusted according to the amount of the remaining outstanding loan of the loan lender n + 1. e (n + 1). In step S208, the personal income of the loan applicant n + 1 is checked, and the credit rating e (n + 1) is adjusted according to the situation. Among them, the personal income of the loan applicant n + 1 during the period of one unit minus the loan repayment is checked. If the difference is less than 1/3 of the individual's income, the credit rating e (n + 1) will be reduced, otherwise the credit rating e (n + 1) will be increased. In step S210, the working company of the loan applicant n + 1 is checked, and the credit rating e (n + 1) is adjusted according to the situation. Among them, the capital or reputation of the working company is checked. If the working company is judged as a good company, then Improve credit rating e (n + 1). In step S212, check the personal expenditure of the credit lender n + 1, and adjust the credit rating e (n + 1) according to the situation. Among them, it is checked whether the credit lender n + 1 has increased a lot of spending recently. If there is no such situation, the credit rating e (n + 1) will be increased. It should be noted that the factors that determine and adjust the credit rating e (n + 1) of the loan applicant n + 1 are not limited to the information listed above, other factors may also be considered. In addition, in some embodiments, one or more of steps S204 to S212 may be selectively omitted. Finally, in step S214, the corresponding default rate c (n + 1) is determined according to the default trend rate of the regional personal credit statistics corresponding to the credit rating e (n + 1) of the credit lender n + 1.

The repayment process detection module 120 is used for receiving instructions from the reserve water level control server 160 to detect the repayment status of each of all the lenders in the database 162 and calculate each of all the lenders The remaining unpaid principal is stored in the database 162. Specifically, please refer to FIG. 3 at the same time. FIG. 3 is a flowchart of a repayment process detection of the loan system 100 disclosed. The repayment process detection module 120 periodically detects the repayment status of all loan applicants in the database 162 according to the instructions of the reserve water level control server 160. In one embodiment, the repayment process detection module 120 detects the repayment status of all loan applicants once a month. As shown in FIG. 3, the repayment process detection module 120 detects repayment status for each lender, for example, if there are n lenders in the lending system 100, i is 1 to n. In step S302, the repayment status of the credit lender i is checked. In step S304, if the credit lender i has repaid in the current period, it proceeds to step S306, sets the number of consecutive unpayment periods k (i) = 0, and sets a default flag d (i) = 0. Then it proceeds to step S314, and calculates the remaining outstanding principal a (i) of the credit lender i. If the credit lender i has not repaid in the current period, the process proceeds to step S308, and the number of consecutive unpaid periods k (i) = k (i) +1 is set. Next, confirm whether the number of consecutive non-payment periods k (i) is small At the predetermined period md, if the number of consecutive non-payment periods k (i) is less than the predetermined period md, proceed to step S312, set the default flag d (i) = 0, and then proceed to step S314 to calculate the remaining amount of the creditor i The principal a (i) was not repaid. If the number of consecutive non-payment periods k (i) is not less than the predetermined number of periods md, the process proceeds to step S316, a default flag d (i) = 1 is set, and the loan lender i is judged to be in default. In some embodiments, the predetermined number of periods md = 6, that is, if the credit lender i has not repaid for more than six consecutive months, the credit lender i is judged to be in default.

When the repayment process detection module 120 detects a default by the credit lender i, it enters a default processing flow. First, the loan system 100 withdraws the amount from the reserve account 170 according to the default payment ratio r (i) of the borrower i in the database 162 when the default occurs, so as to pay the corresponding investors. Then, after paying the investor, the reserve water level detection module 130 receives an instruction from the reserve water level control server 160 and according to the remaining outstanding principal of each of all the creditors remaining in the database 162, The default rate of each of all remaining lenders and the balance of the reserve account 170 are used to calculate the reserve difference D.

In detail, please refer to FIG. 4, which is a flowchart of a reserve water level detection of the loan system 100 of the present disclosure. The reserve water level detection module 130 can calculate the estimated total compensation and the reserve difference D according to the instructions of the reserve water level control server 160 on a regular basis or after a breach occurs and the investor is compensated, and the estimated total compensation and The reserve difference D is stored in the database 162. As shown in FIG. 4, when there are n lenders remaining in the database 162, in step S402, the remaining outstanding principals a (1) to a (n) of each of the lenders 1 to n are respectively set. Multiply by the default rate c (1) ~ c (n) to get the application The lender's estimated default payment of 1 ~ n is possible. Next, in step S404, the estimated default payment of each default of each of the lenders 1 to n is added up, so as to obtain the estimated total payment. Finally, in step S406, the difference between the estimated total compensation and the balance w of the reserve account 170 is calculated so as to obtain a reserve difference D. In one embodiment, formula 1 can be used to calculate the reserve difference D, where c (m) is the default rate, a (m) is the remaining outstanding principal, and w is the balance of the reserve account 170.

The risk allocation ratio control module 140 is configured to receive an instruction from the reserve water level control server 160 and determine the risk allocation ratio of the newly added lender according to the default rate and the reserve balance of the newly added lender. The risk allocation ratio is stored in the database 162. For details, please refer to FIG. 5, which is a flowchart of adding a lender to the loan system 100 of the present disclosure. As shown in FIG. 5, in step S502, the lending system 100 receives a loan application from a new lender n + 1. In step S504, the credit rating detection module 110 determines the credit rating e (n + 1) and default rate c (n + 1) of the credit lender n + 1 based on the data of the credit lender n + 1. Details The calculation method is as described in FIG. 2 above, and is not repeated here. In step S506, the reserve difference D calculated by the reserve water level detection module 130 is obtained. The detailed calculation method is as described in FIG. 4 above, and is not repeated here. In step S508, the risk allocation ratio control module 140 determines the risk allocation ratio p (n + 1) of the lender n + 1 according to the default rate c (n + 1) of the lender n + 1. In step S510, it is determined whether the reserve difference D is greater than 0. If the reserve difference D is not greater than 0, the process proceeds to step S512 to increase the risk allocation ratio p (n + 1). If The reserve difference D is greater than 0, and it proceeds to step S514 to reduce the risk allocation ratio p (n + 1). In one embodiment, if the reserve difference D is 0, the risk allocation ratio p (n + 1) is not adjusted. In one embodiment, formula 2 can be used to calculate the risk allocation ratio p (n + 1), where c (n + 1) is the default rate, D is the reserve difference, and k is the feedback correction coefficient.

p (n + 1) = c ( n +1) × (1+ D × k ) ... Equation 2

In this way, if the balance w of the reserve account 170 is too low, the risk allocation ratio will increase, and if the balance w of the reserve account 170 is too high, the risk allocation ratio will decrease. If the sum of the default rate of all the lenders and the remaining outstanding principal product is low, the risk allocation ratio is reduced. Conversely, if the sum of the default rate of all the lenders and the remaining outstanding principal product is higher , Then increase the risk allocation ratio. By dynamically adjusting the risk appropriation ratio of newly added lenders, the reserve level can be adjusted in real time, thereby maintaining the stable operation of the lending system 100.

Finally, in step S516, according to the credit rating e (n + 1) of the credit lender n + 1, the corresponding default compensation ratio r (n + 1) of the credit lender n + 1 is determined.

The default payout ratio control module 150 is configured to receive an instruction from the reserve water level control server 160 and adjust the default payout ratio of each of all the creditors in the database 162 according to the reserve difference D in the database 162. Please refer to FIG. 6, which is a flowchart of controlling the default payment ratio of the loan system 100 disclosed. The default compensation ratio control module 150 periodically adjusts the default compensation ratio. In one embodiment, the default compensation ratio control module 150 adjusts the default compensation ratio quarterly or semi-annually. Specifically, As shown in FIG. 6, the default compensation ratio control module 150 adjusts the default compensation ratio for each lender, for example, if there are n lenders in the lending system 100, i is 1 to n. In step S602, it is confirmed whether the loan application of the lender i has ended. In step S604, if the loan application of the credit lender i has ended, the default payment ratio r (i) = 0 is set. In step S606, if the loan application of the credit lender i is not completed, it is confirmed whether the reserve difference D is greater than zero. In step S608, if the reserve difference D is greater than 0, the default compensation ratio r (i) is reduced. In step S610, if the reserve difference D is not greater than 0, it is confirmed whether the reserve difference D is less than 0. In step S612, if the reserve difference D is less than 0, the default compensation ratio r (i) is increased. Finally, in step S614, if the reserve difference D is not less than 0, that is, the reserve difference D is equal to 0, the default compensation ratio r (i) does not change. In one embodiment, formula 3 can be used to adjust the default payment ratio r (i), where r (i) is the default payment ratio before adjustment, r '(i) is the adjusted default payment ratio, and D is the reserve difference , S is the feedback correction coefficient, i is 1 to n, and y is the adjustment parameter. In one embodiment, y is 0.07.

r ' ( i ) = r ( i ) × (1- D × s × (1- r ( i )) / y ) ... Equation 3

In this way, if the balance w of the reserve account 170 is too low, the default compensation ratio will be reduced. If the balance w of the reserve account 170 is too high, the default compensation ratio will be increased. If the sum of the default rate of all current lenders and the remaining outstanding principal product becomes low, increase the default compensation ratio. Conversely, if the sum of the default rate of all current lenders and the remaining outstanding principal product becomes higher, Then reduce the default payment ratio. By dynamically adjusting the default payment ratio of all lenders, it is possible to control the level of reserves in real time, thereby maintaining borrowing The stable operation of the loan system 100.

The reserve water level control server 160 periodically sends instructions to the repayment process detection module 120 and the reserve water level detection module 130 to update the reserve difference D in the database 162, and then dynamically adjust the risk increase in the database 162. Allocation ratio and default compensation ratio. . In detail, the reserve water level control server 160 is based on the credit rating detection module 110, the repayment process detection module 120, the reserve water level detection module 130, the risk allocation ratio control module 140, and the default payment. The parameters generated by the proportional control module 150 are coordinated and controlled. When accepting a loan application from a new lender, the risk allocation ratio of the new lender can be dynamically calculated at that time. When a default occurs, the corresponding investor's default compensation ratio at that time can be dynamically determined. Therefore, the loan system 100 can dynamically maintain the reserve water level balance by dynamically adjusting the parameters, so that the default risk is protected by the system's precise feedback calculation.

In one embodiment, the loan system 100 includes an economic prosperity detection module 180 for receiving an instruction from the reserve water level control server 160 to determine the economic prosperity parameter E according to the economic prosperity index at that time, and stored in the database. In 162, the economic prosperity index may be, for example, a prosperity index and a light signal issued by a country. In this embodiment, the risk allocation ratio control module 140 and the default compensation ratio control module 150 further modify the risk allocation ratio and the default compensation ratio according to the economic prosperity parameter E, respectively. In one embodiment, formula 4 and formula 5 can be used to adjust the risk allocation ratio p (n + 1) and the default payment ratio r '(i), where p' (n + 1) is adjusted according to the economic prosperity parameter E The proportion of risk provision, r '(i) is the default compensation ratio adjusted by formula 3, and r "(i) is the default compensation ratio adjusted according to the economic prosperity parameter E, i is 1 to n, , Correct function for feedback.

In one embodiment, the lending system 100 includes a community behavior detection module 190 for receiving instructions from the reserve water level control server 160 and determining the community behavior parameter F based on the community behavior index of the borrower at the time. And stored in the database 162, where the community behavior index can be obtained, for example, by analyzing the frequency and content of the lender's postings on the social networking site. In this embodiment, the risk allocation ratio control module 140 and the default compensation ratio control module 150 further modify the risk allocation ratio and the default compensation ratio according to the community behavior parameter F, respectively. In an embodiment, formula 6 and formula 7 can be used to adjust the risk allocation ratio p '(n + 1) and the default payment ratio r "(i), and p' (n + 1) is the risk adjusted by formula 4. Withdrawal ratio, p "(n + 1) is the risk allocation ratio adjusted according to the community behavior parameter F, r" (i) is the default compensation ratio adjusted by formula 5, and r "'(i) is the basis Adjusted community behavior parameter F, the default compensation ratio, i is 1 to n, , Correct function for feedback.

To sum up, the present invention provides a lending system and a corresponding lending method. As a lending platform applied to the Internet, lenders and investors can freely match lending transactions on this platform. and, Based on the default rate estimated by the credit rating of the borrower, a certain percentage of the amount is allocated to the reserve account. In this way, when the default occurs, the amount in the reserve account can be used to compensate the investor. Certain protection. Due to the fast nature of Internet dissemination, the daily transaction volume is very large, and the lenders need to know their credit rating and their payable costs (proportion of the amount of payment) on this platform in a short period of time. And when investors lend funds to lenders, they also hope to obtain the risk protection of default compensation. The invention disclosed in this disclosure is based on the reserve water level, and dynamically modifies the risk appropriation ratio of the lenders and the default payment ratio of the investors, so that while maintaining the reserve water level, the newly-added lenders can immediately know With regard to its own payable costs, investors can also get system dynamic compensation when a default occurs. The invention disclosed in this disclosure utilizes an automatic feedback correction method to enable the loan system to operate automatically and adjust the parameterization, so that it can cope with a large and fast transaction volume, while dynamically maintaining the stable operation of the loan system.

The following uses examples to further explain the operation of the loan system and loan method of the present disclosure, so that those with ordinary knowledge in the technical field to which the present invention pertains can more easily understand the technical content of the present invention and implement it accordingly. It should be noted that the data in the following examples are for illustrative purposes only, and are not intended to limit the present invention. Those with ordinary knowledge in the technical field should be able to adjust the loan system according to needs without departing from the spirit and scope of the present invention. Parameters.

Adjustment of risk provision ratio and adjustment of default payment ratio

Example 1

In Example 1, suppose a different letter in June 104 Table 1 shows the risk allocation ratio and default compensation ratio corresponding to the rating in the current period.

Assuming a total of 60 million yuan in loan application in the current period, the estimated total compensation is calculated to be 2778,000 yuan based on the estimated default rates of each credit rating, and it is transferred to the reserve account for a total of 2778,000 according to the risk allocation ratio of each credit rating in this period yuan. The total default in this period is 300,000 yuan, and the total compensation in accordance with the default compensation ratio in this period is 162,000 yuan, which is deducted from the reserve account, so the balance of the reserve account is w = 2616000 yuan.

Repayment process detection: the repayment of the principal in this period accounts for the estimated total payment of 0.

Reserve water level detection: The estimated total compensation is 2778,000 yuan, and the system sets the coverage rate to 1.5. Therefore, the reserve water level target is 4167000 yuan, minus reserve account balance w = 2616000 yuan, can get reserve balance D = 1551000.

Risk allocation ratio control: Set the feedback correction coefficient k = 0.0000001, multiply the result by the reserve difference D to 0.1551, and use formula 2 to modify the risk allocation ratio in the next period. For all new loan applications after the amendment, the revised risk allocation ratio applies. In Example 1, because the reserve difference D is greater than 0, which means that the reserve water level is insufficient, the risk allocation ratio in the next period is revised upward.

Default compensation ratio control: Set the feedback correction coefficient s = 0.00000002, multiply the result with the reserve difference D to 0.03102, and use formula 3 to modify the default compensation ratio for the next period. All the defaults that occur after the correction are subject to the revised default compensation ratio. In Example 1, because the reserve difference D is greater than 0, it means that the reserve water level is insufficient, and the default compensation ratio for the next period is revised downward.

After the amendment, the risk allocation ratio and the default compensation ratio in the next period are shown in Table 2.

Example 2

In Example 2, it is assumed that the risk allocation ratio and the default payout ratio for the current period corresponding to different credit ratings at December 104 are shown in Table 3.

Assuming a total of 20 million yuan in loan application in the current period, the estimated total compensation is calculated as 926,000 yuan based on the estimated default rate of each credit rating, and it is transferred to the reserve account according to the risk allocation ratio of each credit rating for a total of 1069622. yuan. The total amount of default in this period is 140,000 yuan, and the total amount of compensation in accordance with the default compensation ratio of this period is 68976 yuan, which is deducted from the reserve account. Therefore, the balance of the reserve account is w = 3616645.

Repayment process detection: the repayment of the principal in this period accounts for the estimated total compensation of 160,000.

Reserve water level detection: The estimated total compensation is 3544,000 yuan, and the system sets the coverage rate to 1.5. Therefore, the reserve water level target is 531600 yuan, minus the reserve account balance w = 3616645 yuan, and the reserve balance D = 1699354 can be obtained. .

Risk allocation ratio control: Set the feedback correction factor k = 0.0000001, multiply the result by the reserve difference D to 0.169935429, and use formula 2 to modify the risk allocation ratio for the next period. For all new loan applications after the amendment, the revised risk allocation ratio applies. In Example 2, because the reserve difference D is greater than 0, which means that the reserve water level is insufficient, the risk allocation ratio in the next period is revised upward.

Default compensation ratio control: Set the feedback correction factor s = 0.00000002, multiply the result with the reserve difference D to 0.033987086, and use formula 3 to modify the default compensation ratio for the next period. All the defaults that occur after the amendment are subject to the revised default compensation ratio. In Example 2, because the reserve difference D is greater than 0, it means that the reserve water level is insufficient, and the default compensation ratio for the next period is revised downward.

After the amendment, the risk allocation ratio and the default payout ratio in the next period are shown in Table 4.

Example 3

In Example 3, it is assumed that the risk allocation ratio and the default payout ratio for the current period corresponding to different credit ratings at June 105 are shown in Table 5.

Assuming a total of 20 million yuan in loan application in the current period, the estimated total compensation will be 926,000 yuan based on the estimated default rate of each credit rating, and the total amount will be transferred to the reserve account according to the credit allocation ratio of the current period to 1083360. yuan. The total amount of default in this period is 20,000 yuan, and the total amount of compensation in accordance with the default compensation ratio in this period is 16034 yuan, which is deducted from the reserve account. Therefore, the balance of the reserve account is w = 4683971 yuan.

Repayment process detection: the repayment of the principal in this period accounts for the estimated total compensation of 1389,000.

Reserve water level detection: The estimated total compensation is 308,100 yuan, and the system sets the coverage rate to 1.5. Therefore, the reserve water level target is 4621500 yuan, minus the reserve account balance w = 4683971 yuan, and the reserve balance D =- 62471.

Risk allocation ratio control: Set the feedback correction coefficient k = 0.0000001, multiply the result by the reserve difference D to -0.006247191, and use formula 2 to correct the risk allocation ratio for the next period. For all new loan applications after the amendment, the revised risk allocation ratio applies. In Example 3, because the reserve difference D is less than 0, it means that the reserve water level is too much, so the risk allocation ratio in the next period is revised downward.

Default compensation ratio control: Set the feedback correction coefficient s = 0.00000002, multiply the result by the reserve difference D to -0.001249438, and use formula 3 to correct the default payment ratio for the next period. All the defaults that occur after the correction are subject to the revised default payment ratio. In Example 3, because the reserve difference D is less than 0, it means that the reserve water level is too much, and the default compensation ratio for the next period is revised upward.

After correction, the risk allocation ratio and the default payout ratio in the next period are shown in Table 6.

Example 4

In Example 4, it is assumed that the risk allocation ratio and the default payout ratio for the current period corresponding to different credit ratings at December 105 are shown in Table 7.

Assuming a total of 20 million yuan in loan application in the current period, the estimated total compensation is calculated as 926,000 yuan based on the estimated default rate of each credit rating, and it is transferred to the reserve account according to the credit allocation ratio of the current period to a total of 920,215. yuan. The total amount of default in this period is 60,000 yuan, and the total amount of compensation in accordance with the default compensation ratio of this period is 31,321 yuan, which is deducted from the reserve account. Therefore, the balance of the reserve account is w = 5572865 yuan.

Repayment process detection: the repayment of the principal in this period accounts for the estimated total compensation of 138,900.

Reserve water level detection: The estimated total compensation is 3868100 yuan, and the system sets the coverage rate to 1.5. Therefore, the reserve water level target is 5802150 yuan, minus the reserve account balance w = 5572865 yuan, and the reserve balance D = 229285 can be obtained. .

Risk allocation ratio control: Set the feedback correction coefficient k = 0.0000001, multiply the result by the reserve difference D to 0.22928455, and use formula 2 to modify the risk allocation ratio for the next period. For all new loan applications after the amendment, the revised risk allocation ratio applies. In Example 4, because the reserve difference D is greater than 0, which means that the reserve water level is insufficient, the risk allocation ratio in the next period is revised upward.

Default payment ratio control: Set the feedback correction coefficient s = 0.00000002, multiply the result by the reserve difference D to 0.004585691, and use formula 3 to correct the default payment ratio for the next period. All the defaults that occur after the correction are subject to the revised default payment ratio. In Example 4, because the reserve difference D is greater than 0, which means that the reserve water level is insufficient, the default payment ratio for the next period is revised downward.

After the amendment, Table 8 shows the risk allocation ratio and default compensation ratio in the next period.

The above description is the best mode for carrying out the present invention. Persons with ordinary knowledge in the technical field to which the present invention pertains should be able to make minor changes, replacements, and replacements without departing from the spirit and architecture of the present invention. The scope of the present invention's rights depends on the scope of the attached patent application.

Claims (20)

A loan system includes: a reserve water level control server with a database, the reserve water level control server is used to receive a loan application from a lender's device, wherein the loan application includes information of the newly added lender ; Credit rating detection module, connected to the reserve water level control server, for receiving the data of the newly added lender from the reserve water level control server, and based on the newly added lender To determine the credit rating and default rate of the new lender, in order to store the credit rating and default rate of the new lender in the database; detection of the repayment process A module connected to the reserve water level control server for receiving instructions from the reserve water level control server to detect the repayment status of each of all loan applicants in the database and calculate all the loan applications The remaining unpaid principal of each person, and the remaining unpaid principal is stored in the database; a reserve water level detection module is connected to the reserve water level control server for receiving from the preparation Jinshui An instruction from the control server to calculate an estimated total compensation and reserve difference based on the remaining outstanding principal and the default rate of each of all the lenders in the database, and the estimated total compensation And the reserve balance are stored in the database; the risk allocation ratio control module is connected to the reserve water level control server for receiving instructions from the reserve water level control server and according to the data in the database The default rate of the newly added lender and the reserve difference determine the risk allocation ratio of the new lender, and the risk allocation ratio of the new lender is stored in the database And the default compensation ratio control module connected to the reserve water level control server for receiving instructions from the reserve water level control server and adjusting all of the data in the database according to the reserve balance of the database. Proportion of default payment for each of the lenders; wherein the reserve water level control server sends instructions to the repayment process detection module and the reserve water level detection module periodically to update The reserve balance in your library, and then dynamically adjusts the risk that the repository set aside the default payout ratio and proportion. The loan system described in item 1 of the scope of patent application, wherein the information of the newly added lender includes the credit report record, repayment record, total loan, personal income, work company and At least one of your personal expenses. The loan system described in item 1 of the scope of patent application, wherein the repayment process detection module detects the repayment status of each of all the lenders and sets the corresponding default flag. When the number of consecutive non-payment periods is less than the predetermined period, the repayment process detection module sets the default flag of the lender to 0, otherwise, sets the default flag of the lender to 1 . The loan system described in item 3 of the scope of patent application, wherein when the repayment process detection module sets the default flag of the borrower to 0, the repayment process detection module recalculates the application The lender's remaining outstanding principal is stored in the database. The loan system described in item 1 of the scope of patent application, wherein the reserve water level detection module is connected to a banking system in order to obtain a reserve account balance, and wherein the reserve water level detection module is based on all the loan applications The remaining unpaid principal and the default rate of each person are used to calculate the estimated total compensation and the reserve difference including: the reserve water level detection module adds up the total of each of the all lenders The result of multiplying the remaining outstanding principal with the default rate in order to obtain the estimated total compensation; and the reserve level detection module calculates the difference between the estimated total compensation and the balance of the reserve account in order to obtain the Reserve balances are stored in the database. The loan system described in item 1 of the scope of patent application, wherein when the reserve difference is greater than zero, the risk allocation ratio control module increases the risk allocation ratio of the newly added lender, and when the reserve When the gold balance is less than zero, the risk allocation ratio control module reduces the risk allocation ratio of the newly added lender. The loan system described in item 1 of the scope of patent application, wherein when the reserve difference is greater than zero, the default payment ratio control module reduces the default payment ratio, and when the reserve difference is less than zero, the default payment ratio The control module increases the default compensation ratio. The loan system described in item 1 of the scope of patent application further includes: an economic prosperity detection module connected to the reserve water level control server for receiving instructions from the reserve water level control server and according to the economic prosperity index The economic prosperity parameter is determined and stored in the database; wherein the risk allocation ratio control module further adjusts the risk allocation ratio of the newly-added lender according to the economic prosperity parameter; and wherein the default payment ratio The control module further adjusts the default payment ratio of each of the all lenders according to the economic prosperity parameter. The loan system described in item 1 of the scope of patent application, further includes: a community behavior detection module connected to the reserve water level control server for receiving instructions from the reserve water level control server and based on all The community behavior index of each of the loan applicants determines the community behavior parameters and stores them in the database; wherein the risk allocation ratio control module further adjusts the newly applied loan according to the community behavior parameters. And the default payment ratio control module further adjusts the default payment ratio of each of all the lenders based on the community behavior parameter. A computer-aided loan method includes the following steps: a reserve water level control server receives a loan application from a lender's device, wherein the loan application includes newly added lender's information; a credit rating detection module The group receives the data of the newly added lender from the reserve water level control server, and determines the credit rating and default rate of the new lender based on the information of the new lender And the credit rating and the default rate of the newly added lender are stored in the database of the reserve water level control server; the risk allocation ratio control module receives the data from the reserve water level control server. Order and based on the default rate and reserve difference of the newly added lender, determine the risk allocation ratio of the new lender, and store the newly allocated lender's risk allocation ratio in In the database; and the default payout ratio control module receives an instruction from the reserve water level control server and determines the default payout of the new lender based on the credit rating of the new lender For example, and the default repayment ratio of the newly added lender is stored in the database; wherein the reserve difference is based on the remaining outstanding principal of each of all the lenders and all the loans Obtained by each person's default rate and reserve account balance; and wherein, by periodically calculating the reserve difference, the risk provision ratio of the newly added lender and all the lenders are dynamically adjusted The default compensation ratio. A loan processing method processed by a computer includes the following steps: The repayment process detection module receives instructions from a reserve water level control server and periodically checks each of all lenders in the database of the reserve water level control server. If one of the all lenders defaults, the reserve water level detection module connected to the reserve water level control server is based on the default lender's loan repayment status in the database. The default compensation ratio pays the corresponding investor; after paying the investor, the reserve water level detection module receives instructions from the reserve water level control server and according to the remaining balance of each of all of the lenders Calculate the reserve balance by failing to repay the principal, the default rate of each of the remaining lenders and the reserve account balance; of which, the newly added lenders are dynamically adjusted by periodically calculating the reserve difference And the proportion of all the lenders' default claims. The borrowing method as described in item 10 of the scope of patent application, wherein the credit rating detection module determines the credit rating of the newly added lender and the default based on the information of the newly added lender. The rate step includes: setting the initial value of the credit rating based on the credit report record of the newly added lender; based on the repayment history, total loan, personal income, work company and individual of the new lender Adjusting at least one of the credit ratings; and determining the default rate of the newly added lender according to the credit rating of the newly added lender. According to the borrowing method described in item 10 of the scope of patent application, the credit rating detection module further performs a credit check on the newly added lender based on the information of the newly added lender. When the newly-added lender does not pass the credit check, the loan application of the newly-added lender is rejected. The method of borrowing as described in item 11 of the scope of patent application, wherein the repayment process detection module periodically checks the repayment status of each of all the lenders to determine whether the default includes the lenders Each: Detect whether the number of consecutive unpaid periods of the lender is greater than the predetermined period. When the number of consecutive unpaid periods is less than the predetermined period, the default flag of the lender is set to 0. Otherwise, the default flag of the lender is set to 1; when the default flag of the lender is 1, the lender is judged to be in default. The loan method described in item 14 of the scope of patent application, wherein when the default flag of the lender is set to 0, the repayment process detection module recalculates the remaining outstanding principal of the lender And store the remaining outstanding principal in the database. The borrowing method described in item 10 or 11 of the scope of patent application, wherein the periodic calculation of the reserve difference includes: the reserve water level detection module receives instructions from the reserve water level control server and adds up all the loan applications A multiplication result of the remaining unpaid principal of each person and the default rate in order to obtain an estimated total compensation and store the estimated total compensation in the database; and the reserve water level detection module The team calculates the difference between the estimated total compensation and the reserve account balance in order to obtain the reserve difference. According to the borrowing method described in item 10 or 11 of the scope of patent application, wherein the periodic calculation of the reserve difference and the dynamic adjustment of the new loan applicant's risk provision ratio include: when the reserve difference is greater than zero, Increase the risk allocation ratio and reduce the risk allocation ratio when the reserve difference is less than zero. The borrowing method as described in item 10 or 11 of the scope of patent application, wherein the periodic calculation of the reserve difference and the dynamic adjustment of the default payment ratios of all the lenders include: when the reserve difference is greater than zero, reducing the default The payout ratio, and when the reserve difference is less than zero, increase the default payout ratio. The borrowing method described in item 10 or 11 of the scope of patent application, further includes: the economic prosperity detection module receives an instruction from the reserve water level control server and determines the economic prosperity parameters according to the economic prosperity index and stores the data in the data And adjust the risk allocation ratio of the newly added lenders and the default compensation ratio of all the lenders according to the economic prosperity parameters. The borrowing method described in item 10 or 11 of the scope of patent application, further comprising: the community behavior detection module receives an instruction from the reserve water level control server and determines based on the community behavior index of all the lenders The community behavior parameters of each of the all lenders are stored in the database, and the risk allocation ratio of the newly added lender and all the lenders are adjusted according to the community behavior parameters The default payout ratio.