Risk prediction method and device, computer equipment and storage medium

Abstract

The invention relates to a risk prediction method and device, computer equipment and a storage medium. The method comprises the following steps: acquiring historical service data of a mobile service from each service platform of a target mechanism, and predicting target service data of the mobile service in a future time period according to the historical service data; determining service classification information of the mobile service according to the historical service data and the target service data; determining mobility index information of the mobility service according to the service classification information; wherein the liquidity index information comprises index values of different liquidity indexes; and determining the mobility risk level of the target mechanism according to the mobility index information and the mobility threshold information. By adopting the method, prospective prediction can be carried out on future data, and the mobility can be monitored in real time.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a risk prediction method, apparatus, computer equipment, and storage medium. Background Technology

[0002] In the new economic environment, improving liquidity risk management is urgently needed. Once a financial institution experiences liquidity risk and incurs losses, it can further impact the entire country's economic order. Currently, most financial institutions' methods for monitoring liquidity risk are relatively outdated. This is due to a lack of professional risk management personnel and the absence of mature stress model theories in the field of liquidity monitoring, hindering the development of effective monitoring methods.

[0003] The existing liquidity risk monitoring process generally only monitors actual data. Manual personnel at various branches periodically summarize and report payment and receipt data in document form. Head office extracts relevant data, prepares reports according to monitoring needs, and then manually queries real-time data from channel systems, core business systems, and credit management systems to identify future liquidity risk points. A final risk report is then generated based on individual experience and a balanced assessment. However, with the continuous growth in the number of systems across different business lines and the volume of customer deposits and withdrawals, as well as the diversification of funding needs, the existing monitoring methods are no longer sufficient to meet the requirements for timely and effective control of liquidity risk. Summary of the Invention

[0004] Therefore, it is necessary to provide a risk prediction method, device, computer equipment, and storage medium to address the aforementioned technical problems, which can make forward-looking predictions of future data and monitor liquidity in real time.

[0005] Firstly, this application provides a risk prediction method, including:

[0006] Historical business data of liquidity services are obtained from various business platforms of the target institution, and target business data of liquidity services in future periods are predicted based on the historical business data.

[0007] Based on the historical business data and the target business data, determine the business classification information of the liquidity business;

[0008] Based on the business classification information, the liquidity indicator information of the liquidity business is determined; wherein, the liquidity indicator information includes the indicator values ​​of different liquidity indicators;

[0009] Based on the liquidity indicator information and liquidity threshold information, the liquidity risk level of the target institution is determined.

[0010] In one embodiment, the classification information includes a first-level classification and a second-level classification located under the first-level classification;

[0011] The step of determining the business classification information of the liquidity business based on the historical business data and the target business data includes:

[0012] Based on the preset value rules for liquidity indicators, the historical business data and the target business data are transformed into at least two standardized liquidity calculation factors.

[0013] By combining different liquidity calculation factors, at least two second-level classifications of the liquidity business can be obtained;

[0014] The liquidity calculation factor and / or at least two second-level classifications are combined to obtain at least two first-level classifications of the liquidity business.

[0015] In one embodiment, determining the liquidity indicator information of the liquidity business based on the business classification information includes:

[0016] For each liquidity indicator, select the target formula model corresponding to the liquidity indicator from the candidate formula models;

[0017] Extract the target classification information required for the target formula model from the business classification information;

[0018] The target classification information is input into the target formula model to obtain the index value of the liquidity index.

[0019] In one embodiment, the liquidity threshold information includes liquidity thresholds corresponding to each liquidity indicator;

[0020] The step of determining the liquidity risk level of the target institution based on the liquidity indicator information and liquidity threshold information includes:

[0021] For any liquidity indicator, the indicator value corresponding to the liquidity indicator is compared with the liquidity threshold, and the target liquidity risk level of the liquidity indicator is determined based on the comparison result.

[0022] The liquidity risk level of the target institution is determined based on the target liquidity risk level of each liquidity indicator.

[0023] In one embodiment, the method further includes:

[0024] Based on the target liquidity risk level of each liquidity indicator, the target liquidity response strategy corresponding to each liquidity indicator is determined from the mapping relationship table; wherein, the mapping relationship table includes the correspondence between the candidate liquidity risk level and the candidate liquidity response strategy for each liquidity indicator;

[0025] Based on the liquidity situation of the target institution in the future period, the various target liquidity response strategies are integrated to obtain the target response strategy for the liquidity business in the future period.

[0026] In one embodiment, the method further includes:

[0027] Based on the target institution's liquidity risk level and target response strategy, generate a risk assessment result;

[0028] The results of the risk assessment are displayed.

[0029] Secondly, this application also provides a risk prediction device, comprising:

[0030] The method further includes:

[0031] Based on the target institution's liquidity risk level and target response strategy, generate a risk assessment result;

[0032] The results of the risk assessment are displayed.

[0033] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0034] Historical business data of liquidity services are obtained from various business platforms of the target institution, and target business data of liquidity services in future periods are predicted based on the historical business data.

[0035] Based on the historical business data and the target business data, determine the business classification information of the liquidity business;

[0036] Based on the business classification information, the liquidity indicator information of the liquidity business is determined; wherein, the liquidity indicator information includes the indicator values ​​of different liquidity indicators;

[0037] Based on the liquidity indicator information and liquidity threshold information, the liquidity risk level of the target institution is determined.

[0038] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the following steps:

[0039] Historical business data of liquidity services are obtained from various business platforms of the target institution, and target business data of liquidity services in future periods are predicted based on the historical business data.

[0040] Based on the historical business data and the target business data, determine the business classification information of the liquidity business;

[0041] Based on the business classification information, the liquidity indicator information of the liquidity business is determined; wherein, the liquidity indicator information includes the indicator values ​​of different liquidity indicators;

[0042] Based on the liquidity indicator information and liquidity threshold information, the liquidity risk level of the target institution is determined.

[0043] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, performs the following steps:

[0044] Historical business data of liquidity services are obtained from various business platforms of the target institution, and target business data of liquidity services in future periods are predicted based on the historical business data.

[0045] Based on the historical business data and the target business data, determine the business classification information of the liquidity business;

[0046] Based on the business classification information, the liquidity indicator information of the liquidity business is determined; wherein, the liquidity indicator information includes the indicator values ​​of different liquidity indicators;

[0047] Based on the liquidity indicator information and liquidity threshold information, the liquidity risk level of the target institution is determined.

[0048] The aforementioned risk prediction method, apparatus, computer equipment, and storage medium acquire historical business data on liquidity operations from various business platforms of the target institution. Based on this historical data, they predict target business data for liquidity operations in future periods. Using both historical and target business data, they determine business classification information for liquidity operations. Based on this classification information, they determine liquidity indicator information, including the values ​​of different liquidity indicators. Finally, based on the liquidity indicator information and liquidity threshold information, they determine the liquidity risk level of the target institution. This solution, by combining historical and target business data, comprehensively analyzes both historical and future data dimensions, avoiding analytical biases caused by missing data. Compared to traditional methods that rely solely on historical data, this approach provides more complete data dimensions. Furthermore, by introducing business classification information, it transforms fragmented business data into structured classification information, resolving the issue of disorganized data determination in traditional methods and ensuring the accuracy of the determined liquidity indicator information, thereby guaranteeing the accuracy of the determined liquidity risk level of the target institution. Attached Figure Description

[0049] To more clearly illustrate the technical solutions in the embodiments or related technologies of this application, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0050] Figure 1 This is a flowchart illustrating a risk prediction method in one embodiment;

[0051] Figure 2 This is a diagram illustrating the structure for collecting historical business data in one embodiment.

[0052] Figure 3 This is a flowchart illustrating the process of determining business classification information in one embodiment;

[0053] Figure 4 This is a schematic diagram of the calculation rules corresponding to the first-level classification in one embodiment;

[0054] Figure 5 This is a flowchart illustrating the process of determining liquidity indicator information for liquidity operations in one embodiment.

[0055] Figure 6 This is a flowchart illustrating the process of determining a target response strategy in one embodiment;

[0056] Figure 7 This is a flowchart illustrating a method in one embodiment;

[0057] Figure 8This is a graph showing the risk assessment results in one embodiment;

[0058] Figure 9 This is a flowchart illustrating the risk prediction method in another embodiment;

[0059] Figure 10 This is a structural block diagram of a risk prediction device in one embodiment;

[0060] Figure 11 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0061] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0062] The risk prediction method provided in this application can be applied to an application environment that manages the liquidity of an institution's assets. The risk prediction method provided in this application can be executed by the server of the target institution's asset management institution (e.g., a finance company), which integrates a liquidity management model to execute the risk prediction method.

[0063] In one exemplary embodiment, such as Figure 1 As shown, a risk prediction method is provided. Taking the application of this method to a server as an example, the specific steps include:

[0064] S101: Obtain historical business data of liquidity business from various business platforms of the target institution, and predict the target business data of liquidity business in future periods based on the historical business data.

[0065] Liquidity operations refer to all business activities undertaken by the target institution's resource management organization around the management of fund liquidity. The core of these activities involves operations such as fund receipt, payment, deposit, lending, and investment, essentially a collection of business activities aimed at balancing fund liquidity, security, and profitability. Business data reflects various data related to fund inflows, outflows, and stock status within a specific period, and is the data foundation for calculating liquidity indicators and assessing liquidity risk. In this embodiment, the target business data refers to the target institution's business data for a future period, which could be the next month. Historical business data refers to the target institution's business data for historical periods.

[0066] Optionally, in this embodiment, the liquidity management module of the target institution's Zichuan management institution (e.g., a finance company) can be connected in real time with business operations such as capital planning, corporate cash inflows and outflows, interbank transactions, and loan disbursements. This enables daily monitoring of key indicators such as the ratio of domestic and foreign currency liquidity, position gaps, interbank lending limits, reserve ratio, and loan-to-deposit ratio to obtain historical business data. For example, relevant data can be automatically extracted from the treasury system's capital planning module, cash management module, and the finance company's settlement system, and a daily cash inflow and outflow table by currency can be established to achieve real-time statistics and monitoring of the finance company's daily cash flow.

[0067] Furthermore, the liquidity monitoring model can use built-in formulas for calculating liquid assets and liabilities to calculate target business data for future periods based on historical business data, such as daily liquidity ratio, position gap, loan-to-deposit ratio, and reserve ratio.

[0068] For example, in the embodiments of this application, a data prediction model can be embedded in the liquidity management module, which can simulate the capital flow of the target institution at the data calculation level to generate target business data.

[0069] In one embodiment, taking an institutional finance company as an example, such as Figure 2 The diagram illustrates a structure for collecting historical business data. The monitoring metrics are liquidity indicators.

[0070] S102, Based on historical business data and target business data, determine the business classification information for liquidity business.

[0071] The classification information includes a first-level classification and second-level classifications under the first-level classification. In this embodiment, the first-level classification can be a summary dimension formed based on the second-level classification, addressing scenarios where the analysis target needs to cover multiple second-level classifications, integrating the scattered second-level classifications into a dimension that fits macro-management needs. For example, the first-level classification is an asset category, and the second-level classifications under this first-level classification may include cash, short-term treasury bonds, interbank certificate of deposit investments, short-term loans, and discounted bills, etc.

[0072] In this embodiment, a two-level classification system can transform fragmented data into analytical dimensions, resolving the issue of matching data scope with management needs. After acquiring historical and target business data, each data type needs to be labeled with key attributes such as data type, business attributes, and time attributes to form a data attribute table. Furthermore, a pre-defined first-level and second-level classification calculation module can be embedded in the liquidity management model. By inputting the labeled historical and target business data into the calculation model, the business classification information of liquidity services can be obtained.

[0073] S103, Based on the business classification information, determine the liquidity indicator information for liquidity business.

[0074] The liquidity indicators include the values ​​of different liquidity indicators. In this embodiment, liquidity indicators include, but are not limited to, the reserve ratio, the 7-day position gap, and the proportion of short-term deposits.

[0075] Optionally, the core liquidity indicators to be calculated can be determined by combining the relevant needs of the target structure and regulatory requirements, avoiding redundancy or omission of key dimensions. After determining the liquidity indicators, the statistical dimensions of each liquidity indicator need to be determined, such as time dimension, subject dimension, currency dimension, etc. Furthermore, the calculation method of each liquidity indicator should be accurately mapped to the business classification information to ensure that the indicator data can be directly extracted from the business classification information or obtained through simple calculations, avoiding data source gaps.

[0076] For example, for each liquidity indicator, the calculation formula can be broken down to clarify which business category information each calculation parameter corresponds to. If the calculation elements of a liquidity indicator cannot directly correspond to business category information, data needs to be extracted from existing categories or supplemented.

[0077] Furthermore, the calculation formula for each liquidity indicator is visualized, specifying the data range and aggregation method to calculate the value of each liquidity indicator. It is important to note that the precision requirements for all liquidity indicator values ​​must be clearly defined in advance; for example, it could be stipulated that all liquidity indicator values ​​be retained to two decimal places. For multi-currency indicators, foreign currency data must be converted to the base currency using the daily midpoint rate before calculation.

[0078] S104. Determine the liquidity risk level of the target institution based on liquidity indicator information and liquidity threshold information.

[0079] The liquidity threshold information may include the overall liquidity threshold of the target institution. Furthermore, it may also include the liquidity thresholds corresponding to each liquidity indicator; each liquidity indicator may have one or more liquidity thresholds. For example, the liquidity thresholds can be set as multiple dimensions such as easing threshold, normal threshold, tight threshold, and crisis threshold, used to rate the risk level of liquidity indicators.

[0080] In this embodiment, if the liquidity threshold information further includes liquidity thresholds corresponding to each liquidity indicator, then the setting of the liquidity threshold corresponding to each liquidity indicator must meet the mandatory regulatory requirements (e.g., a reserve ratio > 3%). The crisis threshold of such liquidity indicators is directly aligned with the regulatory lower limit. Alternatively, the normal threshold can be set to fluctuate around the average of the target structure's indicators over the past 1-2 years.

[0081] Optionally, the overall liquidity situation of the target institution can be analyzed based on liquidity indicator information, and then the liquidity risk level of the target institution can be determined based on the analysis results and liquidity threshold information.

[0082] Furthermore, for any liquidity indicator, the corresponding indicator value and liquidity threshold can be compared, and the target liquidity risk level of the liquidity indicator can be determined based on the comparison results; based on the target liquidity risk level of each liquidity indicator, the liquidity risk level of the target institution can be determined.

[0083] The aforementioned risk prediction method involves obtaining historical business data on liquidity operations from the target institution's various business platforms, and then predicting target business data for liquidity operations in future periods based on this historical data. The method further involves determining business classification information for liquidity operations based on both historical and target business data, then determining liquidity indicator information based on this classification information. This liquidity indicator information includes the values ​​of different liquidity indicators. Finally, the method determines the liquidity risk level of the target institution based on both the liquidity indicator information and liquidity threshold information. This approach, by combining historical and target business data, comprehensively analyzes both historical and future data dimensions, avoiding analytical biases caused by missing data. Compared to the traditional model that only uses historical data, this method provides more complete data dimensions. Furthermore, by introducing business classification information, it transforms fragmented business data into structured classification information, resolving the problem of disorganized data determination in traditional methods, ensuring the accuracy of the determined liquidity indicator information, and consequently, ensuring the accuracy of the determined liquidity risk level of the target institution.

[0084] In one exemplary embodiment, such as Figure 3 As shown, the classification information includes a first-level classification and a second-level classification located under the first-level classification; based on this, a method for determining business classification information is provided, specifically including the following steps:

[0085] S301, based on preset value rules for liquidity indicators, transforms historical business data and target business data into at least two standardized liquidity calculation factors.

[0086] The liquidity calculation factor is a low-level computational unit formed by standardizing the historical and target data of a target institution's liquidity operations according to preset rules (such as data sources and statistical logic). Its core function is to transform scattered raw data (such as deposit balances and loan disbursements) into a unified and reusable basic data module, providing accurate data support for subsequent liquidity business classification and indicator calculation. The preset value rules are a set of clear specifications for data extraction, statistics, and processing, designed in advance to standardize the raw data (historical data / target data) of liquidity operations into a unified liquidity calculation factor. The core is to clarify where to obtain data, how to collect data, and how to process data, ensuring that the factor data extracted by different personnel and at different times is consistent, comparable, and reusable.

[0087] In this embodiment, all liquidity calculation factors must adhere to unified preset rules, including data sources (e.g., only time deposits maturing within 3 months), statistical logic (e.g., average daily balance, cumulative amount), unit (e.g., uniformly in RMB ten thousand), and precision (e.g., retaining two decimal places). This ensures that factors from different time periods and business types are comparable and combinable. For example, the average daily short-term deposit balance factor must explicitly include only demand deposits and time deposits maturing within 3 months, calculated based on the average daily balance over the past year, in RMB ten thousand, and retaining two decimal places. The level of detail in the liquidity calculation factors must align with subsequent analysis needs, being neither too coarse (e.g., only total deposits without breaking down short-term / long-term) nor too fragmented (e.g., broken down to the hourly balance of each deposit). Typically, it focuses on a single business attribute and a single statistical dimension.

[0088] Optionally, based on preset value selection rules, the system clarifies which fields of the target data and / or historical data each liquidity calculation factor data will be extracted from, and determines the statistical method, time range, etc., while also defining a unified data unit and data precision requirements. Further, based on the extracted information, data is automatically extracted from the target data and / or historical data and transformed into at least two standardized liquidity calculation factors.

[0089] S302 combines different liquidity calculation factors to obtain at least two second-level classifications of liquidity business.

[0090] Optionally, based on a pre-defined formulaic model, users can customize the combination calculations between different liquidity calculation factors to obtain at least two second-level classifications of liquidity business. These classifications can be grouped and statistically analyzed by daily or company dimensions. This addresses scenarios where there is a discrepancy between business and analytical dimensions, transforming business operations within the system into analytical dimensions for liquidity analysis. It should be noted that the combined liquidity calculation factors belong to the same business direction, and each second-level classification must correspond to a specific analytical objective, such as assessing the scale of short-term disposable funds for short-term funding sources.

[0091] For example, for each second-level category, a fixed combination rule for liquidity calculation factors is preset. For instance, for the second-level category of daily bank account cash balance, the preset formula is "the sum of the daily balances of all bank accounts"; for the second-level category of daily net capital inflow, the preset formula is "total daily capital inflow minus total daily capital outflow." As another example, the second-level category of short-term deposits requires combining two liquidity calculation factors: the balance of demand deposits maturing within one month and the balance of time deposits maturing between one and three months. It supports statistical breakdown by date (daily, weekly), entity (individual member company, group as a whole), and currency to meet the analytical needs of different levels.

[0092] S303, combine the liquidity calculation factor and / or at least two second-level classifications to obtain at least two first-level classifications for liquidity business.

[0093] Optionally, corresponding to the first-level classification, a custom calculation of liquidity calculation factors and / or combinations of at least two second-level classifications can be performed based on a preset formulaic model to obtain at least two first-level classifications for liquidity business. These can be grouped and statistically analyzed by daily or company dimensions. It should be noted that when a single liquidity indicator needs to cover multiple second-level classifications and / or liquidity calculation factors, they should be aggregated through first-level classifications to avoid dimensional fragmentation during indicator calculation.

[0094] For example, when calculating the total amount of liquid assets, multiple second-level categories such as highly liquid assets and short-term realizable assets need to be integrated. Recombination can be based on both liquidity calculation factors and second-level categories simultaneously, rather than relying solely on a single category. For instance, one of the first-level categories for short-term debt repayment funds can be a combination of deposits maturing within one month (second-level category) and the current day's cash balance (liquidity calculation factor). It should be noted that the calculation of the first-level categories supports customized calculation rules based on aggregation needs; it can be a simple summation or a weighted average. For example, the first-level category of comprehensive liquid assets can be defined as "highly liquid assets × 1 + short-term realizable assets × 0.8" (with weights set according to liquidity strength).

[0095] like Figure 4 The diagram shows the second-level classification, liquidity calculation factor, and calculation rules corresponding to the first-level classification of securities investment where funds can be readily converted into cash, as provided in this embodiment of the application. In this embodiment, the name of the first-level classification can be entered into the corresponding module of the liquidity management system, and the system can then display the corresponding formulaic model.

[0096] In this embodiment, historical data and target data are transformed into standardized liquidity calculation factors by pre-setting value rules, thereby achieving data standardization. Through the progressive combination of factors → second-level classification → first-level classification, a hierarchical classification system of micro-meso-macro is constructed, thereby providing a multi-dimensional analysis perspective. The classification results of different levels can correspond to different management decision-making needs, and the decision-making basis is more sufficient.

[0097] Optionally, in an exemplary embodiment, such as Figure 5 As shown, a method for determining liquidity indicators in liquidity operations is provided, which specifically includes the following steps:

[0098] S501: For each liquidity indicator, select the target formula model corresponding to the liquidity indicator from the candidate formula models.

[0099] It should be noted that, in order to accurately calculate the value of liquidity indicators, a fixed calculation formula can be pre-assigned to each liquidity indicator and stored as a candidate formula model in the liquidity management system. Therefore, in this embodiment, the formula model associated with the liquidity indicator can be selected from the candidate formula models as the target formula model based on the identifier corresponding to the liquidity indicator.

[0100] S502, extract the target classification information required for the target formula model from the business classification information.

[0101] Optionally, first break down the numerator and denominator of the target formula model into specific parameters (e.g., numerator of model C = cash + government bonds + interbank deposits, denominator = various deposits), and then map them one by one to the specific categories in the business classification information. If there are multiple levels of classification, prioritize the second level of classification (fineer granularity, more accurate data). For example, if the numerator of model C requires cash, government bonds, and interbank deposits, you can directly extract the second level of cash classification, government bonds maturing within 7 days classification, and interbank deposits that can be withdrawn at any time classification, rather than the first level of high-liquidity asset classification (if it needs to be split, it will increase the workload).

[0102] S503: Input the target classification information into the target formula model to obtain the index value of the liquidity indicator.

[0103] Optionally, the target classification information can be substituted into the formula of the target formula model in the correct order to avoid errors in the calculation steps (e.g., summing the numerator first, then dividing by the denominator, and finally multiplying by the percentage). Ensure that all input data are in consistent units (e.g., all in ten thousand yuan), and retain decimals according to the preset precision after calculation (e.g., retain two decimal places for index values ​​and one decimal place for percentages). If the input classification data is abnormal (e.g., the denominator is 0 for various deposit balances, or the numerator is negative), it must be processed according to preset rules (e.g., mark the index as invalid, or replace it with the average of the last 3 days) to avoid obtaining meaningless index values.

[0104] For example, regarding the liquidity indicator of the reserve ratio, substituting the target classification information, the numerator = cash classification data (5 million yuan) + government bonds maturing within 7 days classification data (8 million yuan) + interbank deposits available for immediate withdrawal classification data (7 million yuan) = 20 million yuan; the denominator = various deposit classification data (400 million yuan). Calculated according to the target formula model, the reserve ratio = (2000 ÷ 40000) × 100% = 5.0%. Following the rules, one decimal place is retained, and the final indicator value is 5.0%.

[0105] In this embodiment, the closed loop of model matching, classification extraction, and formula calculation avoids the problems of traditional manual number matching and indicator estimation. The indicator values ​​are generated entirely based on structured classification data and fixed models, which reduces the error rate of the calculated liquidity indicator values.

[0106] Optionally, to ensure the stable operation of the target institution, a corresponding response strategy needs to be provided when liquidity risk is detected. In one embodiment, such as... Figure 6 As shown, a method for determining a target response strategy is provided, which specifically includes the following steps:

[0107] S601, based on the target liquidity risk level of each liquidity indicator, determine the target liquidity response strategy corresponding to each liquidity indicator from the mapping relationship table.

[0108] The mapping table includes the correspondence between the candidate liquidity risk level and the candidate liquidity response strategy for each liquidity indicator.

[0109] It should be noted that the mapping table needs to cover core liquidity indicators, all risk levels (loose / normal / tense / crisis) and corresponding candidate strategies. The strategies need to be divided into asset side, liability side, fund allocation and other dimensions to ensure targeting.

[0110] For example, regarding the liquidity indicator of the reserve ratio, if the target liquidity risk level is "tight", the target response strategies extracted from the mapping table are: 1. Redeem 80% of non-core short-term investments; 2. Introduce temporary deposit interest rate hikes; 3. Initiate emergency fund collection.

[0111] S602, based on the liquidity situation of the target institution in the future period, integrate the liquidity response strategies of each target institution to obtain the target response strategy for liquidity business in the future period.

[0112] The liquidity situation of the target institution in the future period is a comprehensive judgment on the balance of supply and demand of funds based on its target business data (such as forecasts of fund inflows / outflows and changes in existing funds) and liquidity indicators.

[0113] Optionally, the overall scenario can be comprehensively judged based on target business data (such as forecast data for the next 1-30 days) and the risk level of a single indicator. Common scenarios and characteristics are as follows: The core characteristics of the tight-loose scenario with short-term tension and medium-term easing are: a position gap of 25 million in the next 7 days (tight), but 120 million treasury bonds will mature in the next 15-30 days (surplus); The core characteristics of the tight-loose scenario with continued tension are: a daily position gap of >10 million in the next 30 days, and a reserve ratio of <5%; The core characteristics of the tight-loose scenario with local tension and overall balance are: a reserve ratio of 4.8% (tight), but a position gap of -8 million in the next 7 days (normal), and a short-term deposit ratio of 35% (normal).

[0114] Furthermore, duplicate measures across different liquidity indicator strategies will be removed to avoid wasting resources, and measures not covered by single liquidity indicator strategies but necessary for the overall scenario will be added. Strategies will be prioritized according to urgency, medium-term, and long-term needs to ensure resources are focused on key measures.

[0115] In this embodiment, strategies for matching single liquidity indicators are first matched, and then the overall scenario is integrated. This not only covers the risk points of each indicator, but also avoids the one-sidedness of strategies caused by a single indicator perspective. When integrating, subsequent measures can be planned in advance by combining future liquidity tightness scenarios, ensuring the comprehensiveness and efficiency of the target response strategy.

[0116] Optionally, in one embodiment, such as Figure 7 As shown, a visualization method is provided, which specifically includes the following steps:

[0117] S701 generates a risk assessment result based on the target institution's liquidity risk level and target response strategy.

[0118] Optionally, information such as the overall risk level results, the details of the risk level of a single liquidity indicator, and the indicator values ​​of the liquidity indicator can be compiled to generate a corresponding table. Furthermore, the corresponding target response strategies can be added to the table to generate a risk assessment result.

[0119] S702 displays the results of the risk assessment.

[0120] Optionally, the display method should be selected based on the usage scenario (such as internal meeting reports, written archives, and real-time system monitoring). The core principles are to highlight key points, make the information intuitive and easy to understand, and ensure completeness. Different colors can be used to mark different risk levels.

[0121] like Figure 8 The diagram shown is a risk assessment result display diagram provided in an embodiment of this application. The corresponding indicator is the asset liquidity ratio. Dates with risks are highlighted in red, and corresponding target response strategies and asset allocation suggestions are generated and displayed.

[0122] In this embodiment, by using structured integration and visualization, the problem of risk levels being described only in words and strategies being listed in a scattered manner is avoided. Decision-makers can quickly grasp the core issues, improving the speed and efficiency of risk response.

[0123] Figure 9 This is a flowchart illustrating a risk prediction method in another embodiment. Based on the above embodiments, this embodiment provides an optional example of a risk prediction method. (Combined with...) Figure 9 The specific implementation process is as follows:

[0124] S901 obtains historical business data of liquidity business from various business platforms of the target institution, and predicts target business data of liquidity business in future periods based on historical business data.

[0125] S902, based on preset value rules for liquidity indicators, transforms historical business data and target business data into at least two standardized liquidity calculation factors.

[0126] S903 combines different liquidity calculation factors to obtain at least two second-level classifications of liquidity business.

[0127] S904, combine the liquidity calculation factor and / or at least two second-level classifications to obtain at least two first-level classifications for liquidity business.

[0128] S905: Select the target formula model corresponding to each liquidity indicator from the candidate formula models.

[0129] S906 extracts the target classification information required for each target formula model from the second-level classification and the first-level classification.

[0130] S907: Input the classification information of each target into the corresponding target formula model to obtain the index values ​​of each liquidity indicator.

[0131] S908 compares the value of each liquidity indicator with the liquidity threshold, and determines the target liquidity risk level of each liquidity indicator based on the comparison results.

[0132] S909 determines the liquidity risk level of a target institution based on the target liquidity risk level of each liquidity indicator.

[0133] S910, based on the target liquidity risk level of each liquidity indicator, determine the target liquidity response strategy corresponding to each liquidity indicator from the mapping relationship table.

[0134] The mapping table includes the correspondence between the candidate liquidity risk level and the candidate liquidity response strategy for each liquidity indicator.

[0135] S911, based on the liquidity situation of the target institution in the future period, integrates the liquidity response strategies of each target institution to obtain the target response strategy for liquidity business in the future period.

[0136] S912 generates a risk assessment result based on the target institution's liquidity risk level and target response strategy, and displays the risk assessment result.

[0137] The specific processes of S901-S912 described above can be found in the description of the above method embodiments. Their implementation principles and technical effects are similar, and will not be repeated here.

[0138] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0139] Based on the same inventive concept, this application also provides a risk prediction device for implementing the risk prediction method described above. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations in one or more risk prediction device embodiments provided below can be found in the limitations of the risk prediction method described above, and will not be repeated here.

[0140] In one exemplary embodiment, such as Figure 10 As shown, a risk prediction device 1000 is provided, including: a data acquisition module 1010, a classification determination module 1020, an indicator determination module 1030, and a risk determination module 1040, wherein:

[0141] The data acquisition module 1010 is used to acquire historical business data of liquidity business from various business platforms of the target institution, and predict the target business data of liquidity business in future periods based on the historical business data.

[0142] The classification determination module 1020 is used to determine the business classification information of liquidity business based on historical business data and target business data.

[0143] The indicator determination module 1030 is used to determine the liquidity indicator information of liquidity business based on business classification information; wherein, the liquidity indicator information includes the indicator values ​​of different liquidity indicators.

[0144] The risk assessment module 1040 is used to determine the liquidity risk level of a target institution based on liquidity indicator information and liquidity threshold information.

[0145] The aforementioned risk prediction device acquires historical business data on liquidity operations from various business platforms of the target institution, and predicts target business data for liquidity operations in future periods based on this historical data. It then determines business classification information for liquidity operations based on both historical and target data. Based on this business classification information, it determines liquidity indicator information for liquidity operations, including the values ​​of different liquidity indicators. Finally, based on the liquidity indicator information and liquidity threshold information, it determines the liquidity risk level of the target institution. This solution, by combining historical and target business data, comprehensively analyzes both historical and future data dimensions, avoiding analytical biases caused by missing data. Compared to the traditional model that only uses historical data, it provides more complete data dimensions. Furthermore, by introducing business classification information, it transforms fragmented business data into structured classification information, solving the problem of disorganized determination of traditional data and ensuring the accuracy of the determined liquidity indicator information, thereby ensuring the accuracy of the determined liquidity risk level of the target institution.

[0146] In one embodiment, the classification information includes a first-level classification and a second-level classification located under the first-level classification; the classification determination module 1020 is specifically used for:

[0147] Based on the preset value rules for liquidity indicators, historical business data and target business data are transformed into at least two standardized liquidity calculation factors; different liquidity calculation factors are combined to obtain at least two second-level classifications of liquidity business; liquidity calculation factors and / or at least two second-level classifications are combined to obtain at least two first-level classifications of liquidity business.

[0148] In one embodiment, the indicator determination module 1030 is specifically used for:

[0149] For each liquidity indicator, select the target formula model corresponding to the liquidity indicator from the candidate formula models; extract the target classification information required by the target formula model from the business classification information; input the target classification information into the target formula model to obtain the indicator value of the liquidity indicator.

[0150] In one embodiment, the liquidity threshold information includes the liquidity thresholds corresponding to each liquidity indicator; the risk determination module 1040 is specifically used for:

[0151] For any liquidity indicator, the corresponding indicator value and liquidity threshold are compared, and the target liquidity risk level of the liquidity indicator is determined based on the comparison results; based on the target liquidity risk level of each liquidity indicator, the liquidity risk level of the target institution is determined.

[0152] In one embodiment, the risk determination module 1040 further includes a strategy determination unit, used for:

[0153] Based on the target liquidity risk level of each liquidity indicator, the target liquidity response strategy corresponding to each liquidity indicator is determined from the mapping relationship table. The mapping relationship table includes the correspondence between the candidate liquidity risk level of each liquidity indicator and the candidate liquidity response strategy. Based on the liquidity tightness of the target institution in the future period, the target liquidity response strategies are integrated to obtain the target response strategy for liquidity business in the future period.

[0154] In one embodiment, the risk determination module 1040 further includes a display unit for:

[0155] Based on the target institution's liquidity risk level and the target response strategy, a risk assessment result is generated; the risk assessment result is then displayed.

[0156] Each module in the aforementioned risk prediction device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the operations corresponding to each module.

[0157] In one exemplary embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 11 As shown, this computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and databases. The internal memory provides the environment for the operating system and computer programs stored in the non-volatile storage media to run. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communicating with external terminals via a network connection. When the computer program is executed by the processor, it implements a risk prediction method.

[0158] Those skilled in the art will understand that Figure 11 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0159] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0160] Obtain historical business data of liquidity business from various business platforms of the target institution, and predict the target business data of liquidity business in future periods based on the historical business data;

[0161] Based on historical business data and target business data, determine the business classification information for liquidity business;

[0162] Based on the business classification information, determine the liquidity indicator information for liquidity business; among which, the liquidity indicator information includes the indicator values ​​of different liquidity indicators;

[0163] The liquidity risk level of the target institution is determined based on liquidity indicator information and liquidity threshold information.

[0164] In one embodiment, the classification information includes a first-level classification and a second-level classification under the first-level classification; when the processor executes the computer program to determine the business classification information of the liquidity business based on historical business data and target business data, it also performs the following steps:

[0165] Based on the preset value rules for liquidity indicators, historical business data and target business data are transformed into at least two standardized liquidity calculation factors; different liquidity calculation factors are combined to obtain at least two second-level classifications of liquidity business; liquidity calculation factors and / or at least two second-level classifications are combined to obtain at least two first-level classifications of liquidity business.

[0166] In one embodiment, when the processor executes a computer program to determine the liquidity indicator information of a liquidity business based on the business classification information, it also performs the following steps:

[0167] For each liquidity indicator, select the target formula model corresponding to the liquidity indicator from the candidate formula models; extract the target classification information required by the target formula model from the business classification information; input the target classification information into the target formula model to obtain the indicator value of the liquidity indicator.

[0168] In one embodiment, the liquidity threshold information includes liquidity thresholds corresponding to each liquidity indicator; when the processor executes the computer program to determine the liquidity risk level of the target institution based on the liquidity indicator information and the liquidity threshold information, it also performs the following steps:

[0169] For any liquidity indicator, the corresponding indicator value and liquidity threshold are compared, and the target liquidity risk level of the liquidity indicator is determined based on the comparison results; based on the target liquidity risk level of each liquidity indicator, the liquidity risk level of the target institution is determined.

[0170] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0171] Based on the target liquidity risk level of each liquidity indicator, the target liquidity response strategy corresponding to each liquidity indicator is determined from the mapping relationship table. The mapping relationship table includes the correspondence between the candidate liquidity risk level of each liquidity indicator and the candidate liquidity response strategy. Based on the liquidity tightness of the target institution in the future period, the target liquidity response strategies are integrated to obtain the target response strategy for liquidity business in the future period.

[0172] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0173] Based on the target institution's liquidity risk level and the target response strategy, a risk assessment result is generated; the risk assessment result is then displayed.

[0174] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor:

[0175] Obtain historical business data of liquidity business from various business platforms of the target institution, and predict the target business data of liquidity business in future periods based on the historical business data;

[0176] Based on historical business data and target business data, determine the business classification information for liquidity business;

[0177] Based on the business classification information, determine the liquidity indicator information for liquidity business; among which, the liquidity indicator information includes the indicator values ​​of different liquidity indicators;

[0178] The liquidity risk level of the target institution is determined based on liquidity indicator information and liquidity threshold information.

[0179] In one embodiment, the classification information includes a first-level classification and a second-level classification under the first-level classification; when the processor executes the computer program to determine the business classification information of the liquidity business based on historical business data and target business data, it also performs the following steps:

[0180] Based on the preset value rules for liquidity indicators, historical business data and target business data are transformed into at least two standardized liquidity calculation factors; different liquidity calculation factors are combined to obtain at least two second-level classifications of liquidity business; liquidity calculation factors and / or at least two second-level classifications are combined to obtain at least two first-level classifications of liquidity business.

[0181] In one embodiment, when the processor executes a computer program to determine the liquidity indicator information of a liquidity business based on the business classification information, it also performs the following steps:

[0182] For each liquidity indicator, select the target formula model corresponding to the liquidity indicator from the candidate formula models; extract the target classification information required by the target formula model from the business classification information; input the target classification information into the target formula model to obtain the indicator value of the liquidity indicator.

[0183] In one embodiment, the liquidity threshold information includes liquidity thresholds corresponding to each liquidity indicator; when the processor executes the computer program to determine the liquidity risk level of the target institution based on the liquidity indicator information and the liquidity threshold information, it also performs the following steps:

[0184] For any liquidity indicator, the corresponding indicator value and liquidity threshold are compared, and the target liquidity risk level of the liquidity indicator is determined based on the comparison results; based on the target liquidity risk level of each liquidity indicator, the liquidity risk level of the target institution is determined.

[0185] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0186] Based on the target liquidity risk level of each liquidity indicator, the target liquidity response strategy corresponding to each liquidity indicator is determined from the mapping relationship table. The mapping relationship table includes the correspondence between the candidate liquidity risk level of each liquidity indicator and the candidate liquidity response strategy. Based on the liquidity tightness of the target institution in the future period, the target liquidity response strategies are integrated to obtain the target response strategy for liquidity business in the future period.

[0187] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0188] Based on the target institution's liquidity risk level and the target response strategy, a risk assessment result is generated; the risk assessment result is then displayed.

[0189] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, performs the following steps:

[0190] Obtain historical business data of liquidity business from various business platforms of the target institution, and predict the target business data of liquidity business in future periods based on the historical business data;

[0191] Based on historical business data and target business data, determine the business classification information for liquidity business;

[0192] Based on the business classification information, determine the liquidity indicator information for liquidity business; among which, the liquidity indicator information includes the indicator values ​​of different liquidity indicators;

[0193] The liquidity risk level of the target institution is determined based on liquidity indicator information and liquidity threshold information.

[0194] In one embodiment, the classification information includes a first-level classification and a second-level classification under the first-level classification; when the processor executes the computer program to determine the business classification information of the liquidity business based on historical business data and target business data, it also performs the following steps:

[0195] Based on the preset value rules for liquidity indicators, historical business data and target business data are transformed into at least two standardized liquidity calculation factors; different liquidity calculation factors are combined to obtain at least two second-level classifications of liquidity business; liquidity calculation factors and / or at least two second-level classifications are combined to obtain at least two first-level classifications of liquidity business.

[0196] In one embodiment, when the processor executes a computer program to determine the liquidity indicator information of a liquidity business based on the business classification information, it also performs the following steps:

[0197] For each liquidity indicator, select the target formula model corresponding to the liquidity indicator from the candidate formula models; extract the target classification information required by the target formula model from the business classification information; input the target classification information into the target formula model to obtain the indicator value of the liquidity indicator.

[0198] In one embodiment, the liquidity threshold information includes liquidity thresholds corresponding to each liquidity indicator; when the processor executes the computer program to determine the liquidity risk level of the target institution based on the liquidity indicator information and the liquidity threshold information, it also performs the following steps:

[0199] For any liquidity indicator, the corresponding indicator value and liquidity threshold are compared, and the target liquidity risk level of the liquidity indicator is determined based on the comparison results; based on the target liquidity risk level of each liquidity indicator, the liquidity risk level of the target institution is determined.

[0200] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0201] Based on the target liquidity risk level of each liquidity indicator, the target liquidity response strategy corresponding to each liquidity indicator is determined from the mapping relationship table. The mapping relationship table includes the correspondence between the candidate liquidity risk level of each liquidity indicator and the candidate liquidity response strategy. Based on the liquidity tightness of the target institution in the future period, the target liquidity response strategies are integrated to obtain the target response strategy for liquidity business in the future period.

[0202] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0203] Based on the target institution's liquidity risk level and the target response strategy, a risk assessment result is generated; the risk assessment result is then displayed.

[0204] It should be noted that the data involved in this application (including but not limited to data used for analysis, data stored, data displayed, etc.) are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0205] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0206] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0207] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A risk prediction method, characterized in that, The method includes: Historical business data of liquidity services are obtained from various business platforms of the target institution, and target business data of liquidity services in future periods are predicted based on the historical business data. Based on the historical business data and the target business data, determine the business classification information of the liquidity business; Based on the business classification information, the liquidity indicator information of the liquidity business is determined; wherein, the liquidity indicator information includes the indicator values ​​of different liquidity indicators; Based on the liquidity indicator information and liquidity threshold information, the liquidity risk level of the target institution is determined.

2. The method according to claim 1, characterized in that, The classification information includes a first-level classification and a second-level classification located under the first-level classification; The step of determining the business classification information of the liquidity business based on the historical business data and the target business data includes: Based on the preset value rules for liquidity indicators, the historical business data and the target business data are transformed into at least two standardized liquidity calculation factors. By combining different liquidity calculation factors, at least two second-level classifications of the liquidity business can be obtained; The liquidity calculation factor and / or at least two second-level classifications are combined to obtain at least two first-level classifications of the liquidity business.

3. The method according to claim 1, characterized in that, The step of determining the liquidity indicator information of the liquidity business based on the business classification information includes: For each liquidity indicator, select the target formula model corresponding to the liquidity indicator from the candidate formula models; Extract the target classification information required for the target formula model from the business classification information; The target classification information is input into the target formula model to obtain the index value of the liquidity index.

4. The method according to claim 1, characterized in that, The liquidity threshold information includes the liquidity thresholds corresponding to each liquidity indicator; The step of determining the liquidity risk level of the target institution based on the liquidity indicator information and liquidity threshold information includes: For any liquidity indicator, the indicator value corresponding to the liquidity indicator is compared with the liquidity threshold, and the target liquidity risk level of the liquidity indicator is determined based on the comparison result. The liquidity risk level of the target institution is determined based on the target liquidity risk level of each liquidity indicator.

5. The method according to claim 4, characterized in that, The method further includes: Based on the target liquidity risk level of each liquidity indicator, the target liquidity response strategy corresponding to each liquidity indicator is determined from the mapping relationship table; wherein, the mapping relationship table includes the correspondence between the candidate liquidity risk level and the candidate liquidity response strategy for each liquidity indicator; Based on the liquidity situation of the target institution in the future period, the liquidity response strategies of each target institution are integrated to obtain the target response strategy for the liquidity business in the future period.

6. The method according to claim 5, characterized in that, The method further includes: Based on the target institution's liquidity risk level and target response strategy, generate a risk assessment result; The results of the risk assessment are displayed.

7. A risk prediction device, characterized in that, The device includes: The data acquisition module is used to acquire historical business data of liquidity business from various business platforms of the target institution, and predict the target business data of liquidity business in future periods based on the historical business data. The classification determination module is used to determine the business classification information of the liquidity business based on the historical business data and the target business data; The indicator determination module is used to determine the liquidity indicator information of the liquidity business based on the business classification information; wherein, the liquidity indicator information includes the indicator values ​​of different liquidity indicators; The risk assessment module is used to determine the liquidity risk level of the target institution based on the liquidity indicator information and liquidity threshold information.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images