CN110826793A

CN110826793A - Value evaluation method, device, electronic equipment and medium for asset allocation

Info

Publication number: CN110826793A
Application number: CN201911051581.0A
Authority: CN
Inventors: 雷健雄
Original assignee: JD Digital Technology Holdings Co Ltd
Current assignee: JD Digital Technology Holdings Co Ltd
Priority date: 2019-10-30
Filing date: 2019-10-30
Publication date: 2020-02-21

Abstract

The present disclosure provides a value evaluation method, a value evaluation apparatus, an electronic device, and a medium for asset allocation. The value evaluation method includes: acquiring a transfer coefficient representing the transfer of the specified service among a plurality of sub-periods in a life cycle to obtain a transfer matrix, wherein the plurality of sub-periods respectively correspond to different estimated object states; acquiring the estimated direct profit of the estimated object in each sub-period at least based on the transfer matrix and the estimated profit of the estimated object in each sub-period; acquiring the estimated direct value of the estimated object in each sub-period based on the estimated direct income of the estimated object in each sub-period and the cost of the estimated object in each sub-period; and acquiring the estimated direct value of the estimated object in the life cycle of the specified service based on the estimated direct value of the estimated object in each sub-cycle, so as to allocate resources for the estimated object based on the estimated direct value of the estimated object in the life cycle of the specified service.

Description

Value evaluation method, device, electronic equipment and medium for asset allocation

Technical Field

The present disclosure relates to the field of computer technologies, and in particular, to a method, an apparatus, an electronic device, and a medium for evaluating a value of asset allocation.

Background

Technical means relating to credit card user lifecycle value (CLV) metering, for example, directly metering the overall value contribution of a single user or a group of users of a particular dimension. For another example, based on the difference of the internal states of the users, including the stages of acquisition, survival and development, the value contribution of a single user at each of the above stages is calculated.

In implementing the disclosed concept, the inventors found that there is at least a problem in the related art that the metering method of the CLV for a single user hardly considers the past transactions and inactive users of the user, and the metering method of the CLV for a user group does not consider the difference between users. The above problems may all cause the accuracy of the estimation result to be low, and further cause the asset allocation result obtained based on the estimation result to fail to meet the user requirement, for example, when the estimation result is inaccurate, the advertisement resource, the service resource, the promotion resource and the like are allocated to the user with low requirement, and the resource waste is caused.

Disclosure of Invention

In view of the above, the present disclosure provides a value evaluation method, a value evaluation apparatus, an electronic device, and a medium, in which the estimation accuracy can meet the user requirement so as to facilitate asset allocation.

One aspect of the present disclosure provides a value assessment method for asset allocation, comprising: firstly, obtaining a transfer coefficient representing the transfer of the specified service among a plurality of sub-periods in a life cycle to obtain a transfer matrix, wherein the plurality of sub-periods respectively correspond to different estimated object states. And then, acquiring the estimated direct profit of the estimated object in each sub-period at least based on the transfer matrix and the estimated profit of the estimated object in each sub-period. And then, acquiring the estimated direct value of the estimated object in each sub-period based on the estimated direct profit of the estimated object in each sub-period and the cost of the estimated object in each sub-period. And then, acquiring the estimated direct value of the estimated object in the life cycle of the specified service based on the estimated direct value of the estimated object in each sub-cycle, so as to allocate resources for the estimated object based on the estimated direct value of the estimated object in the life cycle of the specified service.

According to the embodiment of the disclosure, the benefits of the user can be predicted at different stages of the user based on the long-term value of the user, and the difference among the users is considered, so that the prediction requirements of the user on the values at different stages can be met, the prediction granularity is thinner, and the prediction result is more accurate.

According to an embodiment of the present disclosure, obtaining a transfer coefficient characterizing a transfer of a specified service between a plurality of sub-periods in a lifecycle may include: firstly, the life cycle is divided into a plurality of sub-cycles which do not overlap with each other, wherein the sum of the transfer coefficient transferred to one sub-cycle and the transfer coefficient transferred to the rest of the sub-cycles in the plurality of sub-cycles is 100%. Then, determining the transfer rate of the pre-estimated object for representing the transfer among the plurality of sub-periods in the life cycle of the specified service based on the correlation among the sub-periods, or counting the transfer rate of the pre-estimated object for representing the transfer among the plurality of sub-periods in the life cycle of the specified service based on historical data.

According to the embodiment of the disclosure, obtaining the pre-estimated direct profit of the pre-estimated object in each sub-period at least based on the transfer matrix and the pre-estimated profit of the pre-estimated object in each sub-period comprises: and determining the predicted direct profit of the predicted object in each sub-period based on the transfer matrix, the attrition rate and the predicted profit of the predicted object in each sub-period.

According to the embodiment of the disclosure, the loss rate is obtained through a proportional risk model, and variables of the proportional risk model are related to historical data of the pre-estimated object; the historical data includes at least one of: historical data of consumed money, historical data of repayment, historical data of browsing and historical data of placing an order.

According to the embodiment of the disclosure, determining the predicted direct profit of the predicted object in each sub-period based on the transfer matrix, the attrition rate and the predicted profit of the predicted object in each sub-period comprises: and for any one to-be-processed sub-period in the sub-periods, acquiring a plurality of transfer coefficients transferred to the to-be-processed sub-period from each sub-period in the sub-periods based on the transfer matrix. And then, taking the sum of the transfer coefficients as the profit coefficient of the sub-period to be processed. And then, taking the product of the income coefficient and the loss rate of the to-be-processed sub-period and the estimated income of the to-be-processed sub-period as the estimated direct income of the estimated object in the to-be-processed sub-period.

According to the embodiment of the disclosure, acquiring the estimated direct value of the estimated object in each sub-period based on the estimated direct income of the estimated object in each sub-period and the cost of the estimated object in each sub-period comprises: and processing the estimated direct income of the estimated object in each sub-period and the cost of the estimated object in each sub-period by using a value model to obtain the estimated direct value of the estimated object in each sub-period, wherein the cost of the estimated object in each sub-period is related to the credit line, the use line and the repayment amount.

According to an embodiment of the present disclosure, the life cycle includes a credit card life cycle. Accordingly, the plurality of sub-periods includes: new user acquisition period, inactivity period, transaction period, cycle period, overdue period, default period, and exit period.

According to an embodiment of the present disclosure, the method may further include: firstly, the estimated indirect value of the estimated object in the life cycle is obtained. And then, taking the sum of the estimated direct value and the estimated indirect value of the estimated object in the life cycle as the estimated value of the estimated object in the life cycle.

According to the embodiment of the disclosure, acquiring the estimated indirect value of the estimated object in the life cycle comprises: firstly, determining an ecosphere in which the pre-estimated object is located, wherein the ecosphere comprises a plurality of related objects, and at least one overall incidence relation exists among the related objects. Then, at least one local incidence relation between the pre-estimated object and a plurality of related objects is obtained. Then, the importance of the at least one local incidence relation in the at least one overall incidence relation is evaluated to obtain the importance of the pre-estimated object in the ecosphere. And then, acquiring the estimated indirect value of the estimated object based on the importance of the estimated object in the ecological ring.

According to the embodiment of the disclosure, acquiring the estimated indirect value of the estimated object in the life cycle comprises: firstly, indirect value scoring is carried out on the estimation object based on the attribute information of the estimation object to obtain an indirect value score. Then, a projected indirect value of the projected object is determined based at least on the indirect value score.

Another aspect of the present disclosure provides a value assessment apparatus for asset allocation, comprising: the system comprises a transfer matrix obtaining module, a direct income obtaining module, a direct value obtaining module and a total direct value obtaining module. The transfer matrix obtaining module is used for obtaining a transfer coefficient representing the transfer of the specified service among a plurality of sub-periods in the life cycle so as to obtain a transfer matrix, wherein the plurality of sub-periods respectively correspond to different estimated object states. The direct profit obtaining module is used for obtaining the predicted direct profit of the predicted object in each sub-period at least based on the transfer matrix and the predicted profit of the predicted object in each sub-period. The direct value acquisition module is used for acquiring the estimated direct value of the estimated object in each sub-period based on the estimated direct income of the estimated object in each sub-period and the cost of the estimated object in each sub-period. The total direct value acquisition module is used for acquiring the estimated direct value of the estimated object in the life cycle of the specified service based on the estimated direct value of the estimated object in each sub-cycle so as to allocate resources for the estimated object based on the estimated direct value of the estimated object in the life cycle of the specified service.

According to an embodiment of the present disclosure, the transition matrix obtaining module includes: the device comprises a period dividing unit and a transfer rate acquiring unit. The cycle dividing unit is used for dividing the life cycle into a plurality of sub-cycles which do not overlap with each other, wherein the sum of the transfer coefficient transferred to one sub-cycle and the transfer coefficient transferred to the rest sub-cycles in the plurality of sub-cycles is 100%. The transfer rate obtaining unit is used for determining the transfer rate among the sub-periods based on the correlation among the sub-periods, or counting the transfer rate among the sub-periods based on historical data.

According to the embodiment of the disclosure, the direct profit obtaining module is specifically configured to determine the predicted direct profit of the predicted object in each sub-period based on the transfer matrix, the attrition rate and the predicted profit of the predicted object in each sub-period.

According to an embodiment of the present disclosure, the direct profit acquisition module includes: the device comprises a transfer coefficient acquisition unit, a profit coefficient acquisition unit and an estimation unit. The transfer coefficient acquisition unit is used for acquiring a plurality of transfer coefficients for transferring each sub-period to a sub-period to be processed in the plurality of sub-periods based on the transfer matrix. And the profit coefficient acquisition unit is used for taking the sum of the transfer coefficients as the profit coefficient of the sub-period to be processed. The estimation unit is used for taking the product of the income coefficient and the loss rate of the to-be-processed sub-cycle and the estimated income of the to-be-processed sub-cycle as the estimated direct income of the estimation object in the to-be-processed sub-cycle.

According to the embodiment of the disclosure, the direct value acquisition module is specifically configured to process the estimated direct income of the estimation object in each sub-period and the cost of the estimation object in each sub-period by using the value model to obtain the estimated direct value of the estimation object in each sub-period, wherein the cost of the estimation object in each sub-period is related to the credit limit, the use limit and the repayment amount.

According to an embodiment of the present disclosure, the life cycle includes a credit card life cycle. The plurality of sub-periods includes: new user acquisition period, inactivity period, transaction period, cycle period, overdue period, default period, and exit period.

According to an embodiment of the present disclosure, the apparatus further comprises: an indirect value estimation module and an estimation value acquisition module. The indirect value estimation module is used for acquiring the estimated indirect value of the estimated object in the life cycle. The estimated value acquisition module is used for taking the sum of the estimated direct value and the estimated indirect value of the estimated object in the life cycle as the estimated value of the estimated object in the life cycle.

According to an embodiment of the present disclosure, the indirect value prediction module includes: the system comprises an ecological circle determining unit, an incidence relation obtaining unit, an importance obtaining unit and a first indirect value estimating unit. The ecological circle determining unit is used for determining an ecological circle where the pre-estimated object is located, the ecological circle comprises a plurality of related objects, and at least one overall incidence relation exists among the related objects. The incidence relation obtaining unit is used for obtaining at least one local incidence relation between the pre-estimated object and the plurality of related objects. The importance obtaining unit is used for evaluating the importance of the at least one local incidence relation in the at least one overall incidence relation so as to obtain the importance of the pre-estimated object in the ecological circle. The first indirect value estimation unit is used for acquiring the estimated indirect value of the estimation object based on the importance of the estimation object in the ecological ring.

According to an embodiment of the present disclosure, the indirect value prediction module includes: an indirect value score obtaining unit and a second indirect value estimating unit. The indirect value score obtaining unit is used for carrying out indirect value scoring on the estimation object based on the attribute information of the estimation object to obtain an indirect value score. The second indirect value estimation unit is used for determining the estimated indirect value of the estimation object at least based on the indirect value score.

Another aspect of the present disclosure provides an electronic device comprising one or more processors and a storage, wherein the storage is configured to store executable instructions that, when executed by the processors, implement the method as described above.

Another aspect of the present disclosure provides a computer-readable storage medium storing computer-executable instructions for implementing the method as described above when executed.

Another aspect of the disclosure provides a computer program comprising computer executable instructions for implementing the method as described above when executed.

Drawings

The above and other objects, features and advantages of the present disclosure will become more apparent from the following description of embodiments of the present disclosure with reference to the accompanying drawings, in which:

FIG. 1 schematically illustrates an application scenario of a value assessment method, apparatus, electronic device and medium for asset allocation according to an embodiment of the present disclosure;

FIG. 2 schematically illustrates a system architecture diagram suitable for use with the value assessment method, apparatus, electronic device and medium in accordance with embodiments of the disclosure;

FIG. 3 schematically illustrates a flow chart of a value assessment method for asset allocation according to an embodiment of the present disclosure;

FIG. 4 schematically illustrates a flow chart of credit card usage according to an embodiment of the disclosure;

FIG. 5 schematically illustrates a revenue model diagram according to an embodiment of the disclosure;

FIG. 6 schematically illustrates a revenue model framework in accordance with an embodiment of the disclosure;

FIG. 7 schematically shows a lifecycle partitioning diagram according to an embodiment of the present disclosure;

FIG. 8 schematically shows a schematic diagram of the correlation between cycles according to an embodiment of the present disclosure;

fig. 9 schematically illustrates a schematic of retention according to an embodiment of the present disclosure;

FIG. 10 schematically shows a lifecycle partitioning diagram according to another embodiment of the present disclosure;

FIG. 11 schematically illustrates a schematic diagram of a capital cost structure according to an embodiment of the disclosure;

FIG. 12 schematically illustrates a flow chart of a value assessment method for asset allocation according to another embodiment of the present disclosure;

fig. 13 schematically illustrates a schematic diagram of the significance of an ecosphere according to an embodiment of the present disclosure;

FIG. 14 schematically illustrates a diagram of social importance, according to an embodiment of the disclosure;

FIG. 15 schematically illustrates a block diagram of a value assessment apparatus for asset allocation according to an embodiment of the present disclosure; and

FIG. 16 schematically shows a block diagram of an electronic device according to an embodiment of the disclosure.

Detailed Description

Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings. It should be understood that the description is illustrative only and is not intended to limit the scope of the present disclosure. In the following detailed description, for purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of the embodiments of the disclosure. It may be evident, however, that one or more embodiments may be practiced without these specific details. Moreover, in the following description, descriptions of well-known structures and techniques are omitted so as to not unnecessarily obscure the concepts of the present disclosure.

The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the disclosure. The terms "comprises," "comprising," and the like, as used herein, specify the presence of stated features, steps, operations, and/or components, but do not preclude the presence or addition of one or more other features, steps, operations, or components.

All terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art unless otherwise defined. It is noted that the terms used herein should be interpreted as having a meaning that is consistent with the context of this specification and should not be interpreted in an idealized or overly formal sense.

Where a convention analogous to "at least one of A, B and C, etc." is used, in general such a construction is intended in the sense one having skill in the art would understand the convention (e.g., "a system having at least one of A, B and C" would include but not be limited to systems that have a alone, B alone, C alone, a and B together, a and C together, B and C together, and/or A, B, C together, etc.). Where a convention analogous to "A, B or at least one of C, etc." is used, in general such a construction is intended in the sense one having skill in the art would understand the convention (e.g., "a system having at least one of A, B or C" would include but not be limited to systems that have a alone, B alone, C alone, a and B together, a and C together, B and C together, and/or A, B, C together, etc.). The terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more features.

The value estimation scenario of the credit card user is taken as an example for explanation. The CLV metering method based on different internal state stages of users, including a guest obtaining stage, a retention stage and a development stage can meter respective values in stages and can show the connection among the stages, but the method can only show the overall value of a user group of the stages and cannot show the CLV of a single user as a whole. In addition to the above mentioned disadvantages, current metering methods are based on metering of consumer products and not on the consumer itself. There are certain limitations to credit card users in metering.

For general products, the income is mainly derived from transactions, while for credit cards, most of the income of banks is derived from interest generated by the users cycling loans. But the more the user recycles the bill the greater its risk of breach. The risk of breach has not been taken into account in previous metering methods.

For credit card products, the income contribution of users is realized according to different overdraft and repayment behaviors, the repayment behaviors are further distinguished into situations of entering a cycle stage or overdue and the like along with the purchasing behavior or the withdrawal behavior of users by using credit cards, and the income change is not considered in related metering methods.

The income of the credit card product depends on the used amount and the repayment amount, and the used amount in the t period can be influenced by the repayment condition of the bill amount in the t-1 period due to the limit of the credit amount. Adjustment of the amount for the credit card also affects income, and the related art lacks consideration of the effect.

In summary, given the characteristics of credit card products and the immaturity of current CLV metering methodologies, a set of CLV metering systems needs to be developed specifically for such products, and the present disclosure is directed to exploring a CLV personal user value management framework for users of credit card products. On the basis of perfect historical value analysis, the framework simultaneously deeply digs the potential value of the user, and finally forms a whole set of index analysis and discrimination mechanism.

Embodiments of the present disclosure provide a value assessment method, apparatus, electronic device, and medium for asset allocation. The method comprises a revenue acquisition process and a value estimation process. In the income obtaining process, the transfer coefficient of the pre-estimated object in the life cycle of the appointed service is obtained to obtain a transfer matrix, so that the pre-estimated direct income of the pre-estimated object in each sub-cycle can be obtained at least based on the transfer matrix and the pre-estimated income of the pre-estimated object in each sub-cycle. And after the value acquisition is finished, entering a value estimation process, acquiring the estimated direct value of the estimated object in each sub-period based on the estimated direct income of the estimated object in each sub-period and the cost of the estimated object in each sub-period, and acquiring the estimated direct value of the estimated object in the life cycle of the specified service based on the estimated direct value of the estimated object in each sub-period.

FIG. 1 schematically illustrates an application scenario of a value assessment method, apparatus, electronic device and medium for asset allocation according to an embodiment of the present disclosure.

During a credit card transaction, there are typically five transaction parties, including: card user, credit card issuer, merchant issuer, and union pay. The user uses the credit card to conduct purchasing action at the merchant and then pays the issuer. There are a wide variety of credit cards currently on the market, including secured cards and unsecured cards, etc., through which banks perform various services including credit line extension, cash withdrawal, foreign exchange transactions, etc. The present document only meters for consuming functions in unsecured cards.

When the user is successfully credited, the credit card provides a cyclic loan function within the credit line, which can be continued to be used within the credit line even if the user does not settle the current bill. The main revenue sources for a typical bank include: intermediary business revenue, interest revenue, late fees, annual fees, and others.

While issuing the card to the user, the basic specialization of the credit card has been determined, such as credit line, interest rate, billing period, etc., and the card issuer will receive the first annual fee. When the user begins consuming using the credit card, the bank receives intermediate business revenue. After the billing date, the user may be billed for the credit card and may need to make a payment before the payment date. The user has a plurality of payment modes, such as one-time clearing of the current bill, no one-time clearing of the current bill but payment exceeding the minimum payment amount or even no clearing of the minimum payment amount. The bank income composition is also different for different repayment modes of users. If the user once settles the current bill, the bank can not generate any interest income or late fund, but if the user does not once settle the current bill but exceeds the minimum repayment amount, the user can be defaulted to automatically enter a cycle loan stage, in the cycle loan, the bank can obtain the interest generated by the uncommitted bill amount, once the user does not settle even the minimum repayment amount, the user becomes an overdue user, and for the overdue user, the bank can collect the late fund and additional interest. One point to note here is that the mid-business revenue, interest revenue and late fees occur within a certain billing period, but cash flows only occur when the user makes a payment. Therefore, in a normal billing period, the income of the bank is confirmed with the loan or payment activities at the user. In fig. 1, the specific cash flow is shown by solid lines, and the resulting revenue is shown by dashed lines. If the user has been overdue for more than a certain billing period (e.g., 90 days), the overdue user becomes a default user. For these bill amounts, the bank needs to bear the costs resulting from the default.

The credit line of each user is determined by the bank, and when the used line of the user is more, the available line is less, and the bill amount is higher. When the user uses the credit card, the bank can confirm the income of the intermediate business, and when the user performs the repayment action, the bank can confirm the income of interest, the late deposit and the like according to whether the user enters a circulation state or is overdue. Meanwhile, the consumption and repayment behaviors of the user are not kept unchanged all the time, and may be influenced by the environment, and the behavior habits of the user are changed by factors including the liquidity limit of the user, the economic cycle of the market, the amount adjustment of the bank and the like.

Therefore, it is necessary to consider the above factors to estimate the value of the user at each stage.

As shown in fig. 2, a system architecture 200 according to this embodiment may include a terminal device 201, a network 202, a credit card issuer server 203, a merchant server 204, a merchant issuer server 205, and a unionpay server 206. Network 202 serves as a medium to provide a communication link between terminal device 201, credit card issuer server 203, merchant server 204, merchant issuer server 205, and unionpay server 206. Network 202 may include various connection types, such as wired, wireless communication links, or fiber optic cables, to name a few.

Terminal device 201, credit card issuer server 203, merchant server 204, merchant issuer server 205, and unionpay server 206 perform information interaction via network 202 to receive or send messages, etc. The

terminal devices

201, 202, 203 may have installed thereon various communication client applications, such as a bank-type application, a shopping-type application, a web browser application, a search-type application, an instant messaging tool, a mailbox client, social platform software, etc. (by way of example only).

The

terminal devices

201, 202, 203 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop portable computers, desktop computers, and the like.

Credit card issuer server 203, merchant server 204, merchant issuer server 205, and unionpay server 206 may be servers that provide various services, such as a back office server (for example only). The background management server can analyze and process the received data such as the request and the like, and carry out information interaction based on the processing result.

It should be understood that the number of terminal devices, networks, and servers are merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for implementation. It should be noted that fig. 2 is only an example of a system architecture to which the embodiments of the present disclosure may be applied to help those skilled in the art understand the technical content of the present disclosure, and does not mean that the embodiments of the present disclosure may not be applied to other devices, systems, environments or scenarios.

FIG. 3 schematically illustrates a flow chart of a value assessment method for asset allocation according to an embodiment of the present disclosure.

As shown in fig. 3, the method includes operations S301 to S307.

In operation S301, a transfer coefficient representing transfer of a specified service between a plurality of sub-periods in a life cycle is obtained to obtain a transfer matrix, where the plurality of sub-periods respectively correspond to different pre-estimated object states.

For example, the specified services include credit card services and the like, and the life cycle includes a credit card life cycle. Accordingly, the plurality of sub-periods includes: new user acquisition period, inactivity period, transaction period, cycle period, overdue period, default period, and exit period.

For ease of understanding, the following description will take the enactment service as a credit card as an example.

FIG. 4 schematically shows a flow chart of credit card usage according to an embodiment of the disclosure.

As shown in fig. 4, as the consuming behavior and the repayment behavior of the user continuously occur, the user may transit from one state to another state, which mainly includes the following states: inactive phase (deep sleep phase), transaction phase, cycle phase, overdue phase, default phase, and attrition phase. The specific definitions will be described in detail in the next section. Each stage will ultimately affect the bank's income in a different way. For example: users in the recycling phase contribute interest income, while users in the overdue phase contribute late fees. The bank can influence the income of the final credit card by pricing or adjusting the differentiated rate, the annual fee, the amount and the like of different users or the users. Meanwhile, the bank can adopt a control mechanism or guide the behavior of the user through the characteristics of the credit card.

Fig. 5 schematically illustrates a revenue model diagram according to an embodiment of the disclosure.

The user makes himself in different states through consumption of the credit card and subsequent repayment actions, thereby contributing different income to the bank. Banks generally market incoming cash flows in two ways: firstly, by changing the characteristics of the credit card, including interest rate, credit line, annual fee and the like, wherein the interest rate and the annual fee can directly influence income; secondly, the consumption behavior of the user is regulated and controlled through the characteristics of the credit card, for example, the increase of the credit line can stimulate the consumption of the user, so that the probability of the user entering a cycle stage is increased, and finally, the income is indirectly influenced. It should be noted, however, that such incentives may also increase the probability of a user being overdue and default, which requires a bank to balance revenue against expected losses.

Fig. 6 schematically illustrates a revenue model framework in accordance with an embodiment of the disclosure.

Credit cards are generally divided into five internal states during the user's retention period: inactive phase (deep sleep phase), transaction phase, cycle phase, overdue phase, default phase. In the present embodiment, the definition of each stage is as follows.

Inactive phase (deep sleep phase): if the user has not settled the balance in the last billing cycle and has not consumed anything in the current time, the credit card user is considered to be in an inactive phase in the current time, i.e., the user has neither consumed nor recycled bills. The user does not contribute any revenue at the current time.

A transaction stage: meeting either of the following two conditions may be referred to as a trading phase: firstly, the user completely settles the uncoaled balance of the previous billing period, and secondly, the user has consumption behavior in the current period. Outstanding balances are generally the top balance plus interest due to loans in circulation or late deposits due to overdue behavior. Although interest and late funds will be confirmed during the trading phase, they actually occur during the cycling phase and the overdue phase. The only revenue actually confirmed during the transaction phase is the intermediate business revenue, and is the case when the user has consumed the current time.

A circulation stage: this stage is triggered by the user not clearing the amount of the bill in the period, thus confirming interest income. And the condition that the payment amount of the user needs to meet the minimum payment amount in the last period specified by the bank is called that the user is in the cycle stage. Interest income in the circulation stage generally actually occurs in the payment behavior of the user in the next circulation stage or transaction stage. But at the same time it is also possible that the user's default is not actually obtained. The user in the cycle phase can still continue consuming using the credit card within the available credit limit.

The overdue stage: the stage is triggered because the user does not pay the amount of the bill in the upper period or the payment amount does not reach the minimum payment amount specified by the bank, so that the late-deposit income is confirmed. The late fee may actually occur as a one-time clearing of the user's statement during the late transaction phase, or the reduction may be confirmed as the user violates the default.

A default stage: this stage is triggered by the overdue user not having a payment action, thus confirming the loss.

Fig. 7 schematically illustrates a lifecycle partitioning diagram according to an embodiment of the disclosure. The sub-periods obtained after dividing the life cycle of the credit card can be as shown in fig. 7.

As mentioned above, the bank can influence the behavior of the user by adjusting the credit card characteristics such as interest rate and credit line, and the bank can also influence the income of the bank by the adjustment. For example, changes in interest rates may directly affect revenue.

Specifically, obtaining a transition coefficient characterizing a transition of the specified traffic between the plurality of sub-periods in the lifecycle may comprise the following operations.

Firstly, the life cycle is divided into a plurality of sub-cycles which do not overlap with each other, wherein the sum of the transfer coefficient transferred to one sub-cycle and the transfer coefficient transferred to the rest of the sub-cycles in the plurality of sub-cycles is 100%. The sub-period division can be determined according to the standards of the above stages set by each bank.

Then, determining the transfer rate of the pre-estimated object for representing the transfer among the plurality of sub-periods in the life cycle of the specified service based on the correlation among the sub-periods, or counting the transfer rate of the pre-estimated object for representing the transfer among the plurality of sub-periods in the life cycle of the specified service based on historical data.

FIG. 8 schematically shows a schematic diagram of the correlation between cycles according to an embodiment of the disclosure.

As shown in fig. 8, in one embodiment, the determination of the transition rate based on the correlation is described as an example. While the user is trusted, the user is in an inactive or transaction phase, depending on whether the user has begun using the credit card during the first billing period. Once a user begins consuming using a credit card, the internal state of the user changes according to his/her behavior. Trading stage the user's next stage may transition to a looping stage or an overdue stage, or may continue to keep the trading stage unchanged. Based on the repayment behavior of the user, the circulation phase generally has a high probability that the user still keeps in the circulation phase, and the circulation phase also has a certain probability that the user becomes a transaction phase or an overdue phase. The overdue phase progresses to the end or jumps back to the trading phase or a default. The user of the transaction phase must settle the current bill balance and all phases eventually either switch back to the transaction phase or default. But here the attrition phase is specifically proposed, which represents the clearing of all bills by the user while terminating the credit card business with the bank. The final ending status of the user is churn or default. The transition paths between the sub-periods may be as shown in fig. 8.

And aiming at the prediction of the transfer rate among the sub-periods, recommending that the transfer matrix model is established for each user group after the users are grouped according to the historical consumption and repayment behaviors of the users.

The hop rate between the sub-periods is determined by modeling as follows.

First, the transition rate of all the states of the transition matrix to other states is assumed to depend on the parameter X, and meanwhile, the parameter X is assumed to be a standard normal distribution. The specific process is that assuming the initial stage is G, X is divided into a series of disjoint sub-periods

The probability corresponding to the sub-period is the transition rate from the stage G to the stage G.

And calculating formulas corresponding to the transfer rates of different sub-periods as shown in the formula (1).

P (G, G) represents the transfer rate of stage G to stage G in the historical average transfer matrix. Φ () represents the equation for a standard normal distribution. As can be seen from the figure, the lower limit threshold of the inactive period is ∞, and the upper limit threshold of the overdue period is ∞. The rest quantiles can be calculated by the cumulative probability of the corresponding transfer rate.

Assuming a total of N phases, the initial phase contains N-1 phases, i.e., all phases except the default level. There were N-1 historical average transfer rates per initial stage. Each initial phase can be transferred to N different phases and the probability sums to 1.

Assume that the split point X consists of two components: 1) an independent component Y, which is only relevant to the user's own situation; 2) systematic factor Z, associated with all enterprises. In a general sense, Z measures the credit period, when the economic situation is good, Z is positive, the default rate is low, and more customers are transferred to a transaction stage or a circulation stage from an overdue stage; on the contrary, when the economic situation is poor, Z is negative, the default rate is high, and more customers are shifted to the overdue stage or default stage from the trading stage in a grading way.

Meanwhile, the correlation between X and Z (also called system correlation) is represented by ρ, and X, Y, Z is a standard normal distribution, as shown in equation (2).

And (3) initially calculating the transfer rate, namely when both rho and Z take values of 0, the transfer rate is shown as the formula (3).

The transition matrix for each stage is different from the historical average transition matrix. The formula of the transfer rate after being influenced by the sum of rho and Z is as follows, namely when the values of rho and Z are not 0, the calculation formula of the transfer rate is as shown in formula (4).

And calculating a corresponding transition matrix according to the changed probability sub-period, wherein the transition matrix is a predicted value. After knowing how to calculate the transfer rate according to the system correlation and the systematic factor, the value which can enable the predicted value to be closest to the actual value is calculated through value substitution and derivation. In the prediction process, an industry correlation rho is solved by using a maximum likelihood estimation method, and then a systematic factor Z sequence is calculated by using a least square method. The derivation process of ρ is as follows, and ρ is cycled from 0.01 to 0.99, and the following steps are repeated.

The cumulative probability over the interval-4, 4 due to the standard normal distribution is close to 1. Thus, Z values from-4, -3.99 … 0 … 3.99.99, 4 can be approximately matched for each ρ;

the normal distribution is approximated by the above 800 points with a step size of 0.01 between [ -4, 4 ]: respectively solving corresponding cumulative probabilities □ phi (□), wherein phi is a standard normal distribution cumulative probability function, □ phi (□) is a probability difference value between two adjacent Zs, namely respectively solving □ phi (-3) - □ phi (-4), □ phi (-2) - □ phi (-3) …, □ phi (4) - □ phi (3), wherein the value of Z is-4 to 4, and the probability interval of 99.99 percent is covered;

the formula for calculating the theoretical condition default probability of users in a certain user group under the condition of known ρ and Z may be shown as formula (5).

Assuming that the customer violates the success event and the customer does not violate the failure event, the customer violates the binomial distribution. And (3) calculating the cumulative probability of the clients of the user group in the t period according to the observed number of the clients in the t period and the number of default clients, as shown in the formula (6).

And writing a maximum likelihood function, and deducing the system correlation rho by a maximum likelihood estimation method.

And calculating a corresponding transition matrix according to the changed probability interval, wherein the transition matrix is a predicted value and is compared with the actual transition rate at the current period. The least square method is applied to calculate the minimum MSE, namely the sum of squares of the differences between the predicted value and the actual value. Specifically, the formula is shown in (8).

A time series of Zt, i.e. a time series of the systematic factor Z, is obtained. Finally, a regression model is established aiming at the systematic factors, fitting prediction is carried out by adopting an OLS estimation method, prediction is carried out aiming at the transfer matrix according to the systematic factors, and the obtained transfer matrix can be shown in the table 1.

TABLE 1

When the transfer rate between the sub-periods needs to be determined, the required transfer rate can be directly obtained in a table look-up mode.

In another embodiment, the transfer rate between the sub-periods may also be determined statistically based on historical data. For example, in a T0 period, the number of users in the inactive phase is 100, in a T1 period, 20 users satisfy the condition of the transaction phase, and when the transaction phase is shifted from the inactive phase to the transaction phase, the mobility from the inactive phase to the transaction phase is 20%. The foregoing examples are illustrative only and are not to be construed as limiting the present disclosure.

In operation S303, an estimated direct benefit of the estimated object in each sub-period is obtained based on at least the transfer matrix and the estimated benefit of the estimated object in each sub-period.

The predicted income of the predicted object in each sub-period can adopt the average value of the predicted income of a plurality of predicted objects in each sub-period. Multiple predictors may belong to the same group. For example, multiple pre-estimated objects with similar consumption capabilities form a group.

In one embodiment, for a sub-period, the product of the transfer rate to the sub-period and the predicted benefit of the predictor in each sub-period can be used as the predicted direct benefit of each sub-period. The transfer rate to the sub-period can be obtained by looking up a table. For example, as shown in table 1, the transfer rate to the cycle phase is 0.06+0.5+0.1 — 0.66.

For example, the predicted direct revenue at each sub-period may be measured by the product of the probability of each phase (sub-period) occurring and the revenue prediction for each phase, as shown in equation (9).

In addition to the probability of each stage occurring, revenue prediction for each stage is also a very important part, and can directly affect the accuracy of the final CLV metering. Here, the income is directly influenced and divided into two parts, namely the consumption behavior and the payment behavior of the user, and the prediction of the payment behavior is fully considered in the probability of occurrence of each stage, so that the prediction is mainly carried out on the consumption behavior of the user in each bill period.

Currently, in the inventory management system of retail customers, the management of the life cycle of the customer is very popular, and the life cycle of the customer is also a set of inventory user management system designed on the basis of an RFM model according to the consumption behaviors of the customer. The customer life cycle theory, also called customer relationship life cycle theory, is the whole process from the establishment of business relationship between products and customers to the complete termination of relationship, is the development track of the change of the customer relationship level along with time, and dynamically describes the overall characteristics of the customer relationship in different stages. Reasonably grouping the whole amount of users according to different life cycles of the users, and finally predicting the income of the user group with common consumption behaviors and preference.

Fig. 10 schematically shows a lifecycle partitioning diagram according to another embodiment of the present disclosure.

As shown in fig. 10, the consumption behavior of the user at different stages in the use of the credit card is significantly different, and the curves in fig. 10 represent the variation of the frequency and number of times of consumption by the user. Specifically, according to the characteristics of the bank, the consumption preference of the bank user is analyzed, and the consumption behavior characteristics of the user such as account age, consumption times, consumption amount, last single interval, average consumption interval and the like are generally used for grouping the user in the first-level, second-level and third-level life cycles. Thus, the average spending amount of the group in each sub-period may be used as the spending amount of the user in one billing period.

In another embodiment, the impact of different churn rates for each user on their predicted revenue or transition probabilities may be considered. In particular, prediction of internal state path changes during user retention is an important link in the metering of their CLVs. Because the loss rate of the actual user is not a simple constant, if the loss rate of the constant is used for simply estimating the probability of each stage of the user, a large error is caused, so that the CLV measurement of the user is inaccurate, and if the loss rate of the whole bank is estimated, the difference between different users cannot be reflected. Therefore, in the process of performing CLV metering, it is very important how to predict the loss rate and each internal state of the user more accurately.

The embodiment can adopt survival analysis to model the user's attrition rate aiming at the user's attrition rate, thereby calculating the user's retention time. The jump of the user in different internal states can be considered as the distribution of the survival time, and can be described by using a survival function. The proportional risk model cox model solves this problem well, since the distribution of user retention time is unknown.

In one embodiment, the Proportional Hazard (PH) model has the following properties: the risk functions of different individuals are proportional, i.e. the ratio h (t | x1)/h (t | x2) of the risk functions under the two accompanying variable vectors x1 and x2 does not change with the change in t. The formula of the proportional hazards model is shown in equation (11).

h(t|x)＝h₀(t)exp(β₁x₁+β₂x₂+…+β_px_p) (11)

Wherein x is (x)₁，x₂，……，x_p) β for interpreting variables including at least one of number of consumption, amount of consumption, end-of-order interval, average interval of consumption, the interpreting variables relating to attrition rate of the user_i(i ═ 1, 2, …, p) is a regression coefficient. h is₀(t) is x ═ x₁，x₂，……，x_p) The individual's basic risk function when all are 0.

From the relationship between the risk function and the survival function, the survival function based on the Cox model is shown in equation (12).

S(t)＝S₀(t)exp(β₁x₁+β₂x₂+…+β_px_p)，t＞0 (12)

Wherein S (t) is the cumulative retention rate of the permanently lost clients at the time t β_j(j-1, 2, …, p) is an associated x_j(j ═ 1, 2, …, p). S₀(t) is the reference cumulative retention rate of the permanently lost client at the moment t (i.e. the client cumulative retention rate at the moment t without risk), namely the dynamic retention rate of the client at the period t, and can be obtained by Cox regression analysis.

For example, the time period when the customer retention rate of a typical customer drops to 50% is the average full life cycle time of the customer group in which the customer is located.

In another embodiment, the transition rate may also be obtained by constructing a plurality of time series of transition rates for each user group through historical data, establishing a regression model for each time series, and selecting an interpretation variable related to each time series. And carrying out modeling of a linear regression model. Finally, a model set composed of a plurality of regression models is obtained for each user group.

For example, R_t(trading stage → trading stage) ═ β₀+β₁x₁+β₂x₂+…+β_px_p+ epsilon, … …, Rt (overdue phase → default phase) β₀+β₁x₁+β₂x₂+…+β_px_p+ ε. Wherein x is (x)₁，x₂，……，x_p) To explain the variables. R_tThis may also be done for the transition rate between different sub-periods in the transition matrix.

Fig. 9 schematically shows a schematic of retention according to an embodiment of the present disclosure.

As shown in fig. 9, let T be the required full life cycle time of the subdivided customers, i.e. the time period that the cumulative retention rate of a typical customer drops to 50%. t is t_aIs a lifetime just less than T. t is t_bA lifetime just greater than T; s (t)_a)、S(t_b) Are respectively at t for clients_a、t_bI.e., cumulative customer retention. Then linear interpolation is used as shown in equation (13).

(t_a-t_b)：(t_a-T)＝[S(t_a)-S(t_b)]：[S(t_a)-0.5](13)

The T obtained from the above linear interpolation is the full life cycle time of a typical customer.

Correspondingly, acquiring the pre-estimated direct profit of the pre-estimated object in each sub-period at least based on the transfer matrix and the pre-estimated profit of the pre-estimated object in each sub-period comprises: and determining the estimated direct profit of the estimated object in each sub-period based on the transfer matrix, the attrition rate and the estimated profit of the estimated object in each sub-period, wherein the attrition rate is specific to the estimated object.

Specifically, the loss rate is obtained through a proportional risk model, and variables of the proportional risk model are related to historical data of the pre-estimated object; the historical data includes at least one of: historical data of consumed money, historical data of repayment, historical data of browsing and historical data of placing an order. The life cycle is different because the consumption amount historical data, the payment historical data, the browsing historical data, the order placing historical data and other data of different users are different. For example, a user who consumes a high number of times generally has a longer life cycle than a user who consumes a low number of times. For another example, a user who pays on time generally has a longer life than a user who does not pay on time. Based on these data, the user's attrition rate can be estimated. The proportional risk model may be trained using a training data set. For example, a set of training data (with marked information of the attrition rate) including historical data of the consumption amount, historical data of the payment, historical data of the browsing and historical data of the order placing is input into the proportional risk model, and the output result of the proportional risk model approaches the marked information of the attrition rate by adjusting the parameters of the proportional risk model until a preset condition is reached, such as model convergence or the difference between the output result and the marked information of the attrition rate is less than a preset threshold value.

It should be noted that the churn rate can be related to the transfer rate between the sub-periods, which is interpreted as the user churn occurring during the transfer. The churn rate may be related to the predicted direct benefit, and is interpreted as the loss of users during the transfer process that causes the predicted direct benefit to be impaired, and is not limited herein.

In operation S305, an estimated direct value of the predictor in each sub-period is obtained based on the estimated direct profit of the predictor in each sub-period and the cost of the predictor in each sub-period.

In one embodiment, the estimated direct profit of the estimation object in each sub-period and the cost of the estimation object in each sub-period are processed by using a value model to obtain the estimated direct value of the estimation object in each sub-period.

Specifically, determining the predicted direct benefit of the predicted object in each sub-period based on the transfer matrix, the attrition rate and the predicted benefit of the predicted object in each sub-period comprises, for any one to-be-processed sub-period in the plurality of sub-periods:

firstly, a plurality of transfer coefficients of each sub-period to be transferred to the sub-period to be processed in the plurality of sub-periods are obtained based on the transfer matrix.

And then, taking the sum of the transfer coefficients as the profit coefficient of the sub-period to be processed.

And then, taking the product of the income coefficient and the loss rate of the to-be-processed sub-period and the estimated income of the to-be-processed sub-period as the estimated direct income of the estimated object in the to-be-processed sub-period.

In one embodiment, the method can be implemented by

And calculating the estimated direct value of each sub-period.

Where V is the net gain in a sub-period and δ is the conversion rate, which can be regarded as a cost conversion factor. δ is related to capital, risk and operating costs.

FIG. 11 schematically illustrates a schematic diagram of a capital cost structure according to an embodiment of the disclosure.

As shown in fig. 11, the capital cost may vary due to changes in the market environment and the development and opportunity of the bank itself. If the full life value of a user is to be accurately measured, the capital cost is an indispensable part. At present, the method generally adopted is to establish an internal fund transfer pricing system of a transaction level or a combination level and calculate the fund cost of a user dimension month by month. Meanwhile, the measurement of the user dimension capital cost can also assist the bank in making a differential pricing strategy of the rate for the user.

At present, each bank in China has established a perfect FTP mechanism, so that the measurement of the capital cost of the stock business can be based on the existing FTP mechanism in the row, matched with the FTP price on the daily interest starting date and the latest re-pricing date of each transaction and summarized to the client level. And aiming at the newly added services, each service is matched with the corresponding FTP price when occurring, and then summarized to the user plane.

Due to the wide range of banking and products, credit card users may also be involved with other assets at their issuing bank as retail customers, and if it is desired to be metered more accurately, a combined level of metering may be combined with the financial costs of the retail business involved by the bank and ultimately summarized by user ID.

With respect to risk costs. The expected loss at the debt tier can be calculated based on the bank's current internal rating system, and then the risk cost aggregated to each user. The expected loss is formulated as follows: risk cost (EL) ═ PD × LGD × EAD. Wherein PD is the probability of breach. LGD is a loss due to default. EAD is the total risk window. Since the amount of money involved in the user default is not necessarily all of the amount of money that has not been overdue, such as a part of the amount of money that is subsequently paid, or a part of the amount of default can be offset by means of debt or the like, the LGD needs to be considered. In addition, in order to reduce the risk, when the user has a default behavior, the bank may freeze the credit line, for example, 100 ten thousand credit lines, when a certain condition is triggered, but when the amount related to the default reaches 5 ten thousand, the rest of the credit may be frozen to reduce the possible risk, and therefore, the EAD needs to be considered.

Regarding the operation cost, on the basis of traditional accounting of the cost to a specific use department/organization, the cost generated for the product, channel, customer group and other dimensions can be directly accounted to the corresponding final apportionment object during cost accounting.

For example, a user has a retention time of t, income Rev (t), and capital cost of δ_tThe risk cost is EL_tThe operation cost is EXP (t), and the value contribution V of the user can be obtained according to the formula_t＝Rev(t)-δ_t-EL_t-EXP(t)。

In operation S307, the estimated direct value of the pre-estimated object in the life cycle of the designated service is obtained based on the estimated direct value of the pre-estimated object in each sub-cycle, so as to allocate resources to the pre-estimated object based on the estimated direct value of the pre-estimated object in the life cycle of the designated service.

Wherein the resources include, but are not limited to: advertising resources, serving resources, promotional resources, and the like. For example, the advertisement resources may include whether to push the advertisement and its related information to the user, the push frequency of the advertisement, the strength of the offer, and so on. The service resource may include whether to provide the guest service to the user, such as priority of the service, priority of the user's requirement, and the like. The promotion resource can be whether to push a coupon, a coupon to be paid, trial resource information and the like to the user. Therefore, after the resources are pre-judged to be distributed to the user according to the estimated direct value and/or the estimated indirect value and the like of the user, the resource return rate of the user can effectively improve the utilization rate and the return rate of the resources.

For example, equation (14) can be used to obtain the predicted direct value of the predicted object in the life cycle for the specified service.

Wherein, V_tThe net gain in a sub-period is δ is the discount rate, which can be regarded as a cost correlation coefficient. δ is related to capital, risk and operating costs. LTV is the contribution V of the use discount rate delta to the value of each period_tAnd 6, turning off. By the method, the accuracy of the estimated result of the estimated object in the life cycle and each sub life cycle can be effectively improved.

In another embodiment, to further improve the accuracy of the prediction, the indirect value of the prediction object may be further considered.

FIG. 12 schematically illustrates a flow chart of a value assessment method for asset allocation according to another embodiment of the present disclosure.

As shown in fig. 12, the method may further include operations S1201 to S1203.

In operation S1201, a predicted indirection value of the predicted object in the life cycle is obtained.

Since the form of the user's value is not only monetary but also takes into account non-monetary values, the direct effect on bank revenues caused by the consumption and repayment activities of credit cards discussed in current value calculations, as well as many factors that indirectly affect bank revenues, these values are rarely considered in traditional banks, but these potential values are indispensable components in order to achieve accurate metering of the user's CLV.

Since the issuing agency is typically a bank whose business is usually not limited to just credit card business, the credit card users for each bank are also largely involved in more than one product at the bank, and they contribute their potential value to the bank in view of each user's responsiveness to cross-selling and the user's level and influence in their social circle. The bank has very comprehensive data including fund traffic among individual users, consumption scenes, channels, relatives, contact relationships and the like, so that an ecosphere can be derived based on private fund transfer information, and a customer characteristic label and a calculated social structure index are combined through graph analysis and machine learning.

Fig. 13 schematically illustrates a schematic diagram of the significance of an ecosphere according to an embodiment of the present disclosure. As shown in fig. 13, a credit card user is taken as an example.

The construction of the ecosphere influence model is most important to identify the ecosphere in which each user is located. Generally, the fund association relationship refers to fund flow information between individual accounts, and the fund flow information is easy to obtain in the dimension of a bank. Therefore, the key point of the model is to identify the personal social association relationship, and due to the rise of the internet, the rise of various functions such as fission pull-up, community recommendation and the like, the status and the influence of the individual user in the social ecosphere have a considerable effect on the value measurement.

For various ecological networks of social association relation, such as relatives, friends, colleagues, whether consumption is in the same scene or not, a plurality of ecocircles are constructed by using a graph analysis and machine learning method, influence of users in the ecocircles is scored according to the association relation, key individuals are identified, and the actual contribution value of the users with large influence to a bank is far higher than the current value of the users.

In one embodiment, obtaining the projected indirect value of the projected object over the life cycle may include the following operations.

Firstly, determining an ecosphere in which the pre-estimated object is located, wherein the ecosphere comprises a plurality of related objects, and at least one overall incidence relation exists among the related objects. For example, the at least one overall association includes, but is not limited to, at least one of: a fund association relationship and a social association relationship.

Then, at least one local incidence relation between the pre-estimated object and a plurality of related objects is obtained. For example, a personal funds association and a personal social association are obtained.

Then, the importance of the at least one local incidence relation in the at least one overall incidence relation is evaluated to obtain the importance of the pre-estimated object in the ecosphere.

FIG. 14 schematically shows a diagram of social importance according to an embodiment of the disclosure.

As shown in fig. 14, the ecosphere (indicated by the dashed oval circle) includes a plurality of related objects (indicated by the open circles), wherein there are 4 social objects (having four connecting lines connected thereto) of the two related objects, which are important objects in the ecosphere, and wherein the pre-estimated object is also an important object. The importance of the lines can be determined according to the number of the lines connected with the lines during specific quantization.

And then, acquiring the estimated indirect value of the estimated object based on the importance of the estimated object in the ecological ring.

In operation S1203, the sum of the estimated direct value and the estimated indirect value of the estimation object in the life cycle is used as the estimated value of the estimation object in the life cycle. Therefore, factors considered by the estimation result are more comprehensive, and the accuracy of the estimation result is improved.

In another embodiment, obtaining the projected indirect value of the projected object over the life cycle may include the following operations.

Firstly, indirect value scoring is carried out on the estimation object based on the attribute information of the estimation object to obtain an indirect value score.

Then, a projected indirect value of the projected object is determined based at least on the indirect value score.

The growth of different users can be different from person to person, and the accurate judgment of the growth of the users is also a very important link for measuring the potential value of the users according to the characteristics of the users. Commonly considered dimensions include the user's age, occupation, education level, income level, territory, and the like. For example, the subjective judgment factors are more, so that the estimation objects can be scored by using an expert judgment method in cooperation with the real-time strategy of a bank.

The value evaluation method for asset allocation provided by the embodiment estimates the loss rate of the user by adopting a proportional risk model cox, and meanwhile has a more scientific prediction on the retention time of the user: because the loss rate of the actual user is not a simple constant, if the loss rate of the constant is used for simply estimating the probability of each stage of the user, a large error is caused, so that the CLV measurement of the user is inaccurate, and if the loss rate of the whole bank is estimated, the difference between different users cannot be reflected. The loss rate of each pre-estimated object can be accurately pre-estimated by adopting the proportional risk model cox.

According to the value evaluation method for asset allocation, a survival analysis method is adopted according to the user attrition rate, and the cox model is established on the user attrition rate so as to calculate the retention time of the user. The jump of the user in different internal states can be considered as the distribution of the survival time, and can be described by using a survival function. The proportional risk model cox model solves this problem well, since the distribution of user retention time is unknown.

In the value evaluation method for asset allocation provided by the embodiment, in order to estimate the transfer rate of the user in each stage of the life cycle, a method combining user clustering and a transfer matrix system factor model is applied, and the accuracy of the estimation result is effectively improved by accurately clustering credit card users according to the historical behaviors of the users and the user images of the users.

FIG. 15 schematically illustrates a block diagram of a value valuation apparatus for asset allocation in accordance with an embodiment of the present disclosure.

As shown in fig. 15, the value evaluation apparatus 1500 for asset allocation may include: a transfer matrix acquisition module 1510, a direct revenue acquisition module 1530, a direct value acquisition module 1550, and a total direct value acquisition module 1570.

The transfer matrix obtaining module 1510 is configured to obtain a transfer coefficient representing transfer of a specified service between multiple sub-periods in a life cycle, so as to obtain a transfer matrix, where the multiple sub-periods respectively correspond to different estimated object states.

The direct profit obtaining module 1530 is configured to obtain the predicted direct profit of the predicted object in each sub-period based on at least the transfer matrix and the predicted profit of the predicted object in each sub-period.

The direct value obtaining module 1550 is configured to obtain the estimated direct value of the estimated object in each sub-period based on the estimated direct benefit of the estimated object in each sub-period and the cost of the estimated object in each sub-period.

The total direct value obtaining module 1570 is configured to obtain the estimated direct value of the estimated object in the life cycle of the specified service based on the estimated direct value of the estimated object in each sub-cycle, so as to allocate resources to the estimated object based on the estimated direct value of the estimated object in the life cycle of the specified service.

Specifically, the transition matrix obtaining module 1510 may include: the device comprises a period dividing unit and a transfer rate acquiring unit.

The cycle dividing unit is used for dividing the life cycle into a plurality of sub-cycles which do not overlap with each other, wherein the sum of the transfer coefficient transferred to one sub-cycle and the transfer coefficient transferred to the rest sub-cycles in the plurality of sub-cycles is 100%. The transfer rate obtaining unit is used for determining the transfer rate among the sub-periods based on the correlation among the sub-periods, or counting the transfer rate among the sub-periods based on historical data.

For example, the direct gain acquisition module 1530 is specifically configured to determine the predicted direct gains of the forecasted object in each sub-period based on the transfer matrix, the attrition rate, and the predicted gains of the forecasted object in each sub-period.

In addition, the loss rate is obtained through a proportional risk model, and variables of the proportional risk model are related to historical data of the estimated object; the historical data includes at least one of: historical data of consumed money, historical data of repayment, historical data of browsing and historical data of placing an order.

For another example, the direct benefit acquisition module 1530 may include: the device comprises a transfer coefficient acquisition unit, a profit coefficient acquisition unit and an estimation unit. The transfer coefficient acquisition unit is used for acquiring a plurality of transfer coefficients for transferring each sub-period to a sub-period to be processed in the plurality of sub-periods based on the transfer matrix. And the profit coefficient acquisition unit is used for taking the sum of the transfer coefficients as the profit coefficient of the sub-period to be processed. The estimation unit is used for taking the product of the income coefficient and the loss rate of the to-be-processed sub-cycle and the estimated income of the to-be-processed sub-cycle as the estimated direct income of the estimation object in the to-be-processed sub-cycle.

In one embodiment, the direct value obtaining module 1550 is specifically configured to process the pre-estimated direct profit of the pre-estimated object in each sub-cycle and the cost of the pre-estimated object in each sub-cycle by using a value model to obtain the pre-estimated direct value of the pre-estimated object in each sub-cycle, wherein the cost of the pre-estimated object in each sub-cycle is related to the credit line, the use line and the repayment amount.

In one particular embodiment, the life cycle includes a credit card life cycle. The plurality of sub-periods includes: new user acquisition period, inactivity period, transaction period, cycle period, overdue period, default period, and exit period.

In another embodiment, the apparatus 1500 may further comprise: an indirect value estimation module and an estimation value acquisition module. The indirect value estimation module is used for acquiring the estimated indirect value of the estimated object in the life cycle. The estimated value acquisition module is used for taking the sum of the estimated direct value and the estimated indirect value of the estimated object in the life cycle as the estimated value of the estimated object in the life cycle.

Specifically, the indirect value estimation module may include: the system comprises an ecological circle determining unit, an incidence relation obtaining unit, an importance obtaining unit and a first indirect value estimating unit. The ecological circle determining unit is used for determining an ecological circle where the pre-estimated object is located, the ecological circle comprises a plurality of related objects, and at least one overall incidence relation exists among the related objects. The incidence relation obtaining unit is used for obtaining at least one local incidence relation between the pre-estimated object and the plurality of related objects. The importance obtaining unit is used for evaluating the importance of the at least one local incidence relation in the at least one overall incidence relation so as to obtain the importance of the pre-estimated object in the ecological circle. The first indirect value estimation unit is used for acquiring the estimated indirect value of the estimation object based on the importance of the estimation object in the ecological ring.

For example, the indirect value projection module includes: an indirect value score obtaining unit and a second indirect value estimating unit. The indirect value score obtaining unit is used for carrying out indirect value scoring on the estimation object based on the attribute information of the estimation object to obtain an indirect value score. The second indirect value estimation unit is used for determining the estimated indirect value of the estimation object at least based on the indirect value score.

Any number of modules, sub-modules, units, sub-units, or at least part of the functionality of any number thereof according to embodiments of the present disclosure may be implemented in one module. Any one or more of the modules, sub-modules, units, and sub-units according to the embodiments of the present disclosure may be implemented by being split into a plurality of modules. Any one or more of the modules, sub-modules, units, sub-units according to embodiments of the present disclosure may be implemented at least in part as a hardware circuit, such as a Field Programmable Gate Array (FPGA), a Programmable Logic Array (PLA), a system on a chip, a system on a substrate, a system on a package, an Application Specific Integrated Circuit (ASIC), or may be implemented in any other reasonable manner of hardware or firmware by integrating or packaging a circuit, or in any one of or a suitable combination of software, hardware, and firmware implementations. Alternatively, one or more of the modules, sub-modules, units, sub-units according to embodiments of the disclosure may be at least partially implemented as a computer program module, which when executed may perform the corresponding functions.

For example, any of the transfer matrix acquisition module 1510, the direct profit acquisition module 1530, the direct value acquisition module 1550, and the total direct value acquisition module 1570 may be combined and implemented in one module, or any one of them may be split into a plurality of modules. Alternatively, at least part of the functionality of one or more of these modules may be combined with at least part of the functionality of the other modules and implemented in one module. According to an embodiment of the present disclosure, at least one of the transfer matrix acquisition module 1510, the direct benefit acquisition module 1530, the direct value acquisition module 1550 and the total direct value acquisition module 1570 may be implemented at least in part as a hardware circuit, such as a Field Programmable Gate Array (FPGA), a Programmable Logic Array (PLA), a system on a chip, a system on a substrate, a system on a package, an Application Specific Integrated Circuit (ASIC), or by any other reasonable manner of integrating or packaging a circuit, or by any one of three implementations of software, hardware and firmware, or by any suitable combination of any of them. Alternatively, at least one of the transfer matrix acquisition module 1510, the direct profit acquisition module 1530, the direct value acquisition module 1550 and the overall direct value acquisition module 1570 may be at least partially implemented as computer program modules that, when executed, may perform corresponding functions.

FIG. 16 schematically shows a block diagram of an electronic device according to an embodiment of the disclosure. The electronic device shown in fig. 16 is only an example, and should not bring any limitation to the functions and the scope of use of the embodiments of the present disclosure.

As shown in fig. 16, an electronic device 1600 according to an embodiment of the disclosure includes a processor 1601 that can perform various suitable actions and processes in accordance with a program stored in a Read Only Memory (ROM)1602 or a program loaded from a storage portion 1608 into a Random Access Memory (RAM) 1603. Processor 1601 may include, for example, a general purpose microprocessor (e.g., a CPU), an instruction set processor and/or related chipset and/or a special purpose microprocessor (e.g., an Application Specific Integrated Circuit (ASIC)), or the like. Processor 1601 may also include on-board memory for caching purposes. Processor 1601 may include a single processing unit or multiple processing units for performing different actions of a method flow according to embodiments of the present disclosure.

In the RAM1603, various programs and data necessary for the operation of the system 1600 are stored. The processor 1601, the ROM1602, and the RAM1603 are connected to each other via a bus 1604. Processor 1601 performs various operations of the method flow according to embodiments of the present disclosure by executing programs in ROM1602 and/or RAM 1603. It is to be noted that the program may also be stored in one or more memories other than the ROM1602 and the RAM 1603. The processor 1601 may also perform various operations of the method flows according to embodiments of the present disclosure by executing programs stored in the one or more memories.

In accordance with an embodiment of the present disclosure, the system 1600 may also include an input/output (I/O) interface 1605, the input/output (I/O) interface 1605 also being connected to the bus 1604. The system 1600 may also include one or more of the following components connected to the I/O interface 1605: an input portion 1606 including a keyboard, a mouse, and the like; an output portion 1607 including a display device such as a Cathode Ray Tube (CRT), a Liquid Crystal Display (LCD), and a speaker; a storage portion 1608 including a hard disk and the like; and a communication section 1609 including a network interface card such as a LAN card, a modem, or the like. The communication section 1609 performs communication processing via a network such as the internet. The driver 1610 is also connected to the I/O interface 1605 as needed. A removable medium 1611 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like is mounted on the drive 1610 as necessary, so that a computer program read out therefrom is mounted in the storage portion 1608 as necessary.

According to embodiments of the present disclosure, method flows according to embodiments of the present disclosure may be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program product comprising a computer program embodied on a computer readable storage medium, the computer program containing program code for performing the method illustrated by the flow chart. In such embodiments, the computer program may be downloaded and installed from a network via the communication portion 1609, and/or installed from the removable media 1611. The computer program, when executed by the processor 1601, performs the above-described functions defined in the system of the embodiments of the present disclosure. The systems, devices, apparatuses, modules, units, etc. described above may be implemented by computer program modules according to embodiments of the present disclosure.

The present disclosure also provides a computer-readable storage medium, which may be contained in the apparatus/device/system described in the above embodiments; or may exist separately and not be assembled into the device/apparatus/system. The computer-readable storage medium carries one or more programs which, when executed, implement the method according to an embodiment of the disclosure.

According to embodiments of the present disclosure, the computer-readable storage medium may be a non-volatile computer-readable storage medium, which may include, for example but is not limited to: a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the present disclosure, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. For example, according to embodiments of the present disclosure, a computer-readable storage medium may include one or more memories other than ROM1602 and/or RAM1603 and/or ROM1602 and RAM1603 described above.

The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams or flowchart illustration, and combinations of blocks in the block diagrams or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

Those skilled in the art will appreciate that various combinations and/or combinations of features recited in the various embodiments and/or claims of the present disclosure can be made, even if such combinations or combinations are not expressly recited in the present disclosure. In particular, various combinations and/or combinations of the features recited in the various embodiments and/or claims of the present disclosure may be made without departing from the spirit or teaching of the present disclosure. All such combinations and/or associations are within the scope of the present disclosure.

The embodiments of the present disclosure have been described above. However, these examples are for illustrative purposes only and are not intended to limit the scope of the present disclosure. Although the embodiments are described separately above, this does not mean that the measures in the embodiments cannot be used in advantageous combination. The scope of the disclosure is defined by the appended claims and equivalents thereof. Various alternatives and modifications can be devised by those skilled in the art without departing from the scope of the present disclosure, and such alternatives and modifications are intended to be within the scope of the present disclosure.

Claims

1. A value assessment method for asset allocation, comprising:

obtaining a transfer coefficient representing the transfer of the designated service among a plurality of sub-periods in a life cycle to obtain a transfer matrix, wherein the plurality of sub-periods respectively correspond to different estimated object states;

obtaining the estimated direct profit of the estimated object in each sub-period at least based on the transfer matrix and the estimated profit of the estimated object in each sub-period;

acquiring the estimated direct value of the estimated object in each sub-period based on the estimated direct income of the estimated object in each sub-period and the cost of the estimated object in each sub-period; and

and acquiring the estimated direct value of the estimated object in the life cycle of the specified service based on the estimated direct value of the estimated object in each sub-cycle, so as to allocate resources for the estimated object based on the estimated direct value of the estimated object in the life cycle of the specified service.

2. The method of claim 1, wherein the obtaining transition coefficients characterizing a transition of the specified traffic between the plurality of sub-periods in the lifecycle comprises:

dividing the life cycle into a plurality of sub-cycles which do not overlap with each other, wherein the sum of the transfer coefficient transferred to one sub-cycle and the transfer coefficients transferred to the rest of the sub-cycles in the plurality of sub-cycles is 100%;

and determining the transfer rate of the pre-estimated object for representing the transfer among the plurality of sub-periods in the life cycle of the specified service based on the correlation among the sub-periods, or counting the transfer rate of the pre-estimated object for representing the transfer among the plurality of sub-periods in the life cycle of the specified service based on historical data.

3. The method of claim 1, wherein the obtaining the predicted direct benefit of the predictor in each sub-period based on at least the transfer matrix and the predicted benefit of the predictor in each sub-period comprises:

and determining the estimated direct profit of the estimated object in each sub-period based on the transfer matrix, the attrition rate and the estimated profit of the estimated object in each sub-period, wherein the attrition rate is specific to the estimated object and is related to at least one of consumption times, consumption amount, last single interval and average consumption interval.

4. The method according to claim 3, wherein the attrition rate is obtained by a proportional risk model, and variables of the proportional risk model are related to historical data of a pre-estimated object; the historical data includes at least one of: historical data of consumed money, historical data of repayment, historical data of browsing and historical data of placing an order.

5. The method of claim 3, wherein said determining the projected direct return of the predictor for each sub-period based on the transition matrix, attrition rate, and projected return of the predictor for each sub-period comprises: for any pending sub-cycle of the plurality of sub-cycles,

obtaining a plurality of transfer coefficients transferred to the to-be-processed sub-cycle from each sub-cycle in the plurality of sub-cycles based on the transfer matrix;

taking the sum of the transfer coefficients as a profit coefficient of the sub-period to be processed; and

and taking the product of the income coefficient of the to-be-processed sub-period, the loss rate and the estimated income of the to-be-processed sub-period as the estimated direct income of the estimated object in the to-be-processed sub-period.

6. The method of claim 1, wherein said deriving an estimated direct value of said predictor for each sub-period based on an estimated direct benefit of said predictor for each sub-period and a cost of said predictor for each sub-period comprises:

and processing the estimated direct income of the estimated object in each sub-period and the cost of the estimated object in each sub-period by using a value model to obtain the estimated direct value of the estimated object in each sub-period.

7. The method of claim 1, wherein,

the life cycle comprises a credit card life cycle;

the plurality of sub-periods includes: new user acquisition period, inactivity period, transaction period, cycle period, overdue period, default period, and exit period.

8. The method of claim 1, further comprising:

acquiring the estimated indirect value of the estimated object in the life cycle; and

and taking the sum of the estimated direct value and the estimated indirect value of the estimated object in the life cycle as the estimated value of the estimated object in the life cycle.

9. The method of claim 7, wherein said obtaining a projected indirect value of said subject over a life cycle comprises:

determining an ecosphere in which the pre-estimated object is located, wherein the ecosphere comprises a plurality of related objects, and at least one overall association relationship exists among the related objects;

acquiring at least one local incidence relation between the pre-estimated object and the plurality of related objects;

evaluating the importance of the at least one local association relation in the at least one overall association relation to obtain the importance of the pre-estimated object in the ecosphere; and

and acquiring the estimated indirect value of the estimated object based on the importance of the estimated object in the ecological ring.

10. The method of claim 7, wherein said obtaining a projected indirect value of said subject over a life cycle comprises:

indirect value scoring is carried out on the pre-estimated object based on the attribute information of the pre-estimated object, and an indirect value score is obtained; and

determining a projected indirect value of the predictor based at least on the indirect value score.

11. A value assessment apparatus for asset allocation comprising:

the system comprises a transfer matrix obtaining module, a prediction module and a prediction module, wherein the transfer matrix obtaining module is used for obtaining a transfer coefficient for representing the transfer of the specified service among a plurality of sub-periods in a life cycle so as to obtain a transfer matrix, and the plurality of sub-periods respectively correspond to different predicted object states;

the direct profit obtaining module is used for obtaining the predicted direct profit of the predicted object in each sub-period at least based on the transfer matrix and the predicted profit of the predicted object in each sub-period;

the direct value acquisition module is used for acquiring the estimated direct value of the estimated object in each sub-period based on the estimated direct income of the estimated object in each sub-period and the cost of the estimated object in each sub-period; and

and the total direct value acquisition module is used for acquiring the estimated direct value of the estimated object in the life cycle of the specified service based on the estimated direct value of the estimated object in each sub-cycle so as to allocate resources to the estimated object based on the estimated direct value of the estimated object in the life cycle of the specified service.

12. The apparatus of claim 10, wherein the transition matrix obtaining module comprises:

the cycle dividing unit is used for dividing the life cycle into a plurality of sub-cycles which do not overlap with each other, wherein the sum of the transfer coefficient transferred to one sub-cycle and the transfer coefficients transferred to the rest sub-cycles in the plurality of sub-cycles is 100%; and

and the transfer rate acquisition unit is used for determining the transfer rate among the sub-periods based on the correlation among the sub-periods or counting the transfer rate among the sub-periods based on historical data.

13. An electronic device, comprising:

one or more processors;

a storage device for storing executable instructions which, when executed by the processor, implement the method of any one of claims 1 to 10.

14. A computer readable storage medium having stored thereon executable instructions which, when executed by a processor, implement a method according to any one of claims 1 to 10.