A method and device for early warning of asset-liability ratio, electronic equipment and storage medium

Abstract

The application provides a warning method and device for asset-liability ratio, electronic equipment and storage medium. The warning method comprises the following steps: based on the asset-liability data set corresponding to a target enterprise, an asset-liability index system corresponding to the target enterprise is constructed; based on the asset-liability index system, a multiple prediction method is used in combination with external adjustment indication information to obtain a predicted value corresponding to each financial index at a future time point; based on the predicted value, a pre-trained multiple linear regression model is used for prediction to obtain a predicted asset-liability ratio of the target enterprise at the future time point; the predicted asset-liability ratio is compared with a pre-configured warning threshold to obtain a comparison result, and the target enterprise is warned about the asset-liability ratio according to the comparison result. Through the above method, the effect of predicting the asset-liability ratio based on comprehensive data and financial indicators is realized, and the accuracy and efficiency of warning the asset-liability ratio of the enterprise are improved.

Description

Technical Field

[0001] This application relates to the field of financial data processing technology, and in particular to a method, apparatus, electronic device and storage medium for early warning of debt-to-equity ratio. Background Technology

[0002] The relevant opinions and requirements clearly state that by establishing and improving the asset-liability constraint mechanism for state-owned enterprises and strengthening supervision and management, the asset-liability ratio of highly indebted state-owned enterprises should be brought back to a reasonable level as soon as possible. In this process, intelligent prediction and monitoring of the asset-liability ratio, and the ability to efficiently and accurately estimate and detect the asset-liability ratio, are crucial for the operation of enterprises.

[0003] Currently, existing methods for early warning of debt-to-asset ratio mainly involve using self-built financial systems to query debts and perform data calculations. For example, one can view financial statements in a self-built financial system or check a company's loan status through a bank. However, the accuracy of debt or debt-to-asset ratio early warnings using self-built financial systems is low, and reviewing a company's accounting books is time-consuming. In addition, the information provided by banks is not comprehensive enough, resulting in untimely dynamic responses to real-time warnings, which reduces the accuracy and efficiency of debt and debt-to-asset ratio early warnings. Summary of the Invention

[0004] In view of this, the purpose of this application is to provide a method, device, electronic device and storage medium for early warning of debt-to-asset ratio. By constructing a debt-to-asset indicator system for the target enterprise and determining the financial indicators under the system, the predicted value of each financial indicator at a future time point is predicted. Then, a multiple linear regression model is used to predict the debt-to-asset ratio. By comparing the predicted debt-to-asset ratio with the early warning threshold, an early warning of the debt-to-asset ratio of the target enterprise is achieved. This realizes the effect of predicting the debt-to-asset ratio based on comprehensive data and financial indicators, and improves the accuracy and efficiency of early warning of the debt-to-asset ratio of enterprises.

[0005] This application provides an early warning method for the debt-to-asset ratio, the early warning method including:

[0006] Based on a pre-defined asset and liability dataset corresponding to the target enterprise, an asset and liability indicator system corresponding to the target enterprise is constructed; wherein, the indicator system includes multiple indicator dimensions and at least one financial indicator corresponding to each indicator dimension, as well as the indicator calculation formula and data source for each financial indicator;

[0007] Based on the aforementioned asset and liability indicator system, using a preset multiple prediction method and combined with external adjustment indication information, the predicted value of each financial indicator at a preset future time point is obtained.

[0008] Based on the predicted values, a pre-trained multiple linear regression model is used to predict the target company's projected debt-to-equity ratio at the future time point.

[0009] The predicted debt-to-asset ratio is compared with a pre-configured early warning threshold to obtain a comparison result. Based on the comparison result, an early warning is issued to the target company regarding its debt-to-asset ratio.

[0010] Furthermore, the step of constructing an asset and liability indicator system for the target enterprise based on a pre-defined asset and liability dataset includes:

[0011] Based on the preset asset and liability dataset corresponding to the target enterprise, and combined with the preset accounting dimensions and project data, the basic financial data corresponding to the asset and liability dataset is determined.

[0012] The basic financial data is cleaned to obtain the indicator system data corresponding to the basic financial data, and the mapping relationship of the basic financial data is determined.

[0013] Based on the data of the indicator system and the mapping relationship, combined with the preset debt risk quantitative assessment system and the operating characteristics of the target enterprise, an asset and liability indicator system corresponding to the target enterprise is constructed.

[0014] Furthermore, based on the asset-liability indicator system, the predicted value of each financial indicator at a preset future time point is obtained using a preset multiple prediction method and combined with external adjustment indication information, including:

[0015] Based on the data source corresponding to each of the financial indicators, the actual value of the financial indicator at the current point in time is obtained using the indicator calculation formula.

[0016] Based on the actual values, the predicted multiple value corresponding to the financial indicator is obtained using a preset multiple prediction method.

[0017] In response to receiving an external adjustment instruction to adjust the predicted multiple value, the predicted multiple value is adjusted based on the external adjustment instruction and the future time point corresponding to the preset prediction period, so as to obtain the predicted value of the financial indicator at the future time point.

[0018] Furthermore, the multiple linear regression model is pre-trained through the following steps:

[0019] The financial indicators are determined as the independent variables of the multiple linear regression model, and the debt-to-equity ratio is determined as the dependent variable of the multiple linear regression model.

[0020] Based on the asset and liability dataset and the asset and liability indicator system, the corresponding linear relationship between the independent variable and the dependent variable is determined, and the calculation expression of the multiple linear regression model is obtained to obtain the multiple linear regression model to be trained.

[0021] The preset residual method is used to check whether the multivariate linear regression model to be trained satisfies the assumptions corresponding to the multivariate linear regression model.

[0022] If the conditions are met, feature selection and feature construction are performed on the independent variable, and the direction of influence of the independent variable is determined.

[0023] Based on the asset and liability dataset and the direction of influence, the multiple linear regression model to be trained is trained using a preset training method, and the performance of the multiple linear regression model to be trained is optimized to obtain the weight parameters corresponding to the calculation expression, so as to obtain the trained multiple linear regression model.

[0024] Furthermore, the step of using a pre-trained multiple linear regression model to predict the target company's projected debt-to-equity ratio at the future time point based on the predicted value includes:

[0025] Based on the weight parameters and the operating conditions of the target enterprise, the actual weight parameters are determined through a preset scientific calculation model.

[0026] Based on the predicted values ​​and the actual weight parameters, the predicted debt-to-asset ratio of the target enterprise at the future time point is obtained using the calculation expression in the multiple linear regression model.

[0027] Furthermore, before comparing the debt-to-equity ratio with a pre-configured warning threshold, the method further includes:

[0028] Based on the predicted debt-to-asset ratio and the historical debt-to-asset ratio of the target company obtained from the preset historical dataset, the backtest value of the debt-to-asset ratio of the target company is determined, and the predicted debt-to-asset ratio, the historical debt-to-asset ratio, and the backtest value of the debt-to-asset ratio are visualized externally.

[0029] This application embodiment also provides an early warning device for the debt-to-asset ratio, the early warning device comprising:

[0030] The indicator system module is used to construct an asset and liability indicator system for the target enterprise based on a preset asset and liability dataset. The indicator system includes multiple indicator dimensions and at least one financial indicator corresponding to each indicator dimension, as well as the indicator calculation formula and data source for each financial indicator.

[0031] The first prediction module is used to obtain the predicted value of each financial indicator at a preset future time point based on the asset and liability indicator system, using a preset multiple prediction method and combined with external adjustment indication information.

[0032] The second prediction module is used to make predictions based on the predicted values ​​using a pre-trained multiple linear regression model, and to obtain the predicted debt-to-asset ratio of the target enterprise at the future time point.

[0033] The comparison and early warning module is used to compare the predicted debt-to-asset ratio with a pre-configured early warning threshold, obtain the comparison result, and issue an early warning to the target company regarding the debt-to-asset ratio based on the comparison result.

[0034] Furthermore, when the indicator system module is used to construct the asset and liability indicator system corresponding to the target enterprise based on the preset asset and liability dataset corresponding to the target enterprise, the indicator system module is used for:

[0035] Based on the preset asset and liability dataset corresponding to the target enterprise, and combined with the preset accounting dimensions and project data, the basic financial data corresponding to the asset and liability dataset is determined.

[0036] The basic financial data is cleaned to obtain the indicator system data corresponding to the basic financial data, and the mapping relationship of the basic financial data is determined.

[0037] Based on the data of the indicator system and the mapping relationship, combined with the preset debt risk quantitative assessment system and the operating characteristics of the target enterprise, an asset and liability indicator system corresponding to the target enterprise is constructed.

[0038] Furthermore, when the first prediction module is used to obtain the predicted value of each financial indicator at a preset future time point based on the asset-liability indicator system, using a preset multiple prediction method and combining external adjustment indication information, the first prediction module is used to:

[0039] Based on the data source corresponding to each of the financial indicators, the actual value of the financial indicator at the current point in time is obtained using the indicator calculation formula.

[0040] Based on the actual values, the predicted multiple value corresponding to the financial indicator is obtained using a preset multiple prediction method.

[0041] In response to receiving an external adjustment instruction to adjust the predicted multiple value, the predicted multiple value is adjusted based on the external adjustment instruction and the future time point corresponding to the preset prediction period, so as to obtain the predicted value of the financial indicator at the future time point.

[0042] Furthermore, when the second prediction module is used to pre-train the multiple linear regression model, the second prediction module is used to:

[0043] The financial indicators are determined as the independent variables of the multiple linear regression model, and the debt-to-equity ratio is determined as the dependent variable of the multiple linear regression model.

[0044] Based on the asset and liability dataset and the asset and liability indicator system, the corresponding linear relationship between the independent variable and the dependent variable is determined, and the calculation expression of the multiple linear regression model is obtained to obtain the multiple linear regression model to be trained.

[0045] The preset residual method is used to check whether the multivariate linear regression model to be trained satisfies the assumptions corresponding to the multivariate linear regression model.

[0046] If the conditions are met, feature selection and feature construction are performed on the independent variable, and the direction of influence of the independent variable is determined.

[0047] Based on the asset and liability dataset and the direction of influence, the multiple linear regression model to be trained is trained using a preset training method, and the performance of the multiple linear regression model to be trained is optimized to obtain the weight parameters corresponding to the calculation expression, so as to obtain the trained multiple linear regression model.

[0048] Furthermore, when the second prediction module is used to predict the target company's projected debt-to-equity ratio at the future time point based on the predicted value using a pre-trained multiple linear regression model, the second prediction module is used to:

[0049] Based on the weight parameters and the operating conditions of the target enterprise, the actual weight parameters are determined through a preset scientific calculation model.

[0050] Based on the predicted values ​​and the actual weight parameters, the predicted debt-to-asset ratio of the target enterprise at the future time point is obtained using the calculation expression in the multiple linear regression model.

[0051] Furthermore, the early warning device also includes a data display module, which is used for:

[0052] Based on the predicted debt-to-asset ratio and the historical debt-to-asset ratio of the target company obtained from the preset historical dataset, the backtest value of the debt-to-asset ratio of the target company is determined, and the predicted debt-to-asset ratio, the historical debt-to-asset ratio, and the backtest value of the debt-to-asset ratio are visualized externally.

[0053] This application also provides an electronic device, including: a processor, a memory, and a bus. The memory stores machine-readable instructions executable by the processor. When the electronic device is running, the processor communicates with the memory via the bus. When the machine-readable instructions are executed by the processor, the steps of the debt-to-equity ratio early warning method described above are performed.

[0054] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, performs the steps of the aforementioned debt-to-equity ratio early warning method.

[0055] The asset-liability ratio early warning method, device, electronic device, and storage medium provided in this application embodiment include: constructing an asset-liability indicator system for the target enterprise based on a preset asset-liability dataset; wherein the indicator system includes multiple indicator dimensions and at least one financial indicator corresponding to each indicator dimension, as well as an indicator calculation formula and data source for each financial indicator; based on the asset-liability indicator system, using a preset multiple prediction method and combined with external adjustment indication information, obtaining a predicted value for each financial indicator at a preset future time point; based on the predicted value, using a pre-trained multiple linear regression model to predict the predicted asset-liability ratio of the target enterprise at the future time point; comparing the predicted asset-liability ratio with a pre-configured early warning threshold to obtain a comparison result, and issuing an early warning regarding the asset-liability ratio of the target enterprise based on the comparison result.

[0056] Compared to existing methods that rely on self-built financial systems for debt inquiries and data calculations, this new approach constructs an asset-liability indicator system for the target company, determines the financial indicators within that system, predicts the future value of each financial indicator, and then uses a multiple linear regression model to predict the debt-to-asset ratio. By comparing the predicted debt-to-asset ratio with a warning threshold, it provides early warnings about the target company's debt-to-asset ratio. This approach achieves the effect of predicting the debt-to-asset ratio based on comprehensive data and financial indicators, improving the accuracy and efficiency of early warnings regarding the company's debt-to-asset ratio.

[0057] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0058] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0059] Figure 1 One of the flowcharts for an early warning method for the debt-to-equity ratio provided in this application embodiment;

[0060] Figure 2 An example diagram illustrating the prediction of a financial indicator of net working capital provided for an embodiment of this application;

[0061] Figure 3 A schematic diagram illustrating the relationship between net working capital and cash flow statement provided for an embodiment of this application;

[0062] Figure 4 A second flowchart illustrating an early warning method for the debt-to-equity ratio provided in this application embodiment;

[0063] Figure 5 This is one of the structural schematic diagrams of an early warning device for the debt-to-asset ratio provided in an embodiment of this application;

[0064] Figure 6 A second schematic diagram of the structure of an early warning device for the debt-to-asset ratio provided in an embodiment of this application;

[0065] Figure 7 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0066] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of this application. Based on the embodiments of this application, every other embodiment obtained by those skilled in the art without inventive effort falls within the scope of protection of this application.

[0067] Research has found that current methods for early warning of debt-to-equity ratios mainly rely on self-built financial systems to query debts and perform data calculations. For example, this involves viewing financial statements within the self-built financial system or verifying a company's loan status through a bank. While these methods may solve the problem of companies querying debts and calculating debt-to-equity ratios, the accuracy of debt or debt-to-equity ratio early warnings through self-built financial systems is relatively low, and reviewing the company's accounting books is time-consuming. Furthermore, the information provided by banks is not comprehensive enough, resulting in untimely dynamic responses to real-time warnings, thus reducing the accuracy and efficiency of debt and debt-to-equity ratio early warnings.

[0069] Based on this, this application provides a method for early warning of debt-to-asset ratio. By constructing a debt-to-asset indicator system for the target enterprise and determining the financial indicators under this system, the predicted value of each financial indicator at a future time point is predicted. Then, a multiple linear regression model is used to predict the debt-to-asset ratio. By comparing the predicted debt-to-asset ratio with the early warning threshold, an early warning is given to the target enterprise regarding its debt-to-asset ratio. This achieves the effect of predicting the debt-to-asset ratio based on comprehensive data and financial indicators, and improves the accuracy and efficiency of early warning of the enterprise's debt-to-asset ratio.

[0070] In summary, traditional early warning methods are relatively mature in calculating the current debt-to-equity ratio based on financial statements and accounting records. However, for more complex debt-to-equity ratio forecasting, especially involving sensitivity analysis of the ratio, more intelligent solutions are needed. Intelligent forecasting requires early identification and early warning, which necessitates the construction of a risk monitoring and early warning mechanism to monitor key debt risk indicators and conduct real-time tracking and analysis. Furthermore, intelligent forecasting should reference a debt risk quantitative assessment system and consider the company's actual operating environment and conditions. By establishing a personalized corporate debt risk quantitative assessment mechanism, comprehensively considering asset and liability structure, debt level, profitability, cash flow, and asset quality, accurate identification of corporate debt risk can be achieved.

[0071] Please see Figure 1 , Figure 1 This is one of the flowcharts for an early warning method for the debt-to-equity ratio provided in an embodiment of this application. For example... Figure 1 As shown in the embodiments of this application, the method for early warning of the debt-to-asset ratio includes:

[0072] S101. Based on the preset asset and liability dataset corresponding to the target enterprise, construct the asset and liability indicator system corresponding to the target enterprise; wherein, the indicator system includes multiple indicator dimensions and at least one financial indicator corresponding to each indicator dimension, as well as the indicator calculation formula and data source corresponding to each financial indicator.

[0073] It should be noted that the prediction of debt-to-equity ratios for various companies in the energy sector involves complex and numerous characteristics. Clearly defining these characteristics significantly aids in training the model and predicting the debt-to-equity ratio. For example, the ratio of long-term to short-term debt helps determine solvency and debt maturity; high interest rates and short repayment periods can create financial pressure on companies, directly impacting the debt-to-equity ratio; and determining the ratio of shareholders' equity to debt within the capital structure helps understand the proportion of external financing to equity capital.

[0074] Therefore, when using a multiple linear regression model to train on general financial data or to predict the debt-to-equity ratio in the energy industry, the model suffers from overfitting due to insufficient sample size. While it can accurately predict historical results, it may not accurately predict future outcomes. Consequently, the performance of the multiple linear regression model is poor, and its accuracy in predicting the debt-to-equity ratio is not high. To address this challenge, a debt-to-equity dataset is pre-defined for target companies in the energy industry. The data in this dataset is then stratified to construct a corresponding debt-to-equity indicator system for the target companies, thereby improving the accuracy of debt-to-equity ratio prediction.

[0075] The data sources for the asset and liability dataset may include, but are not limited to, the balance sheet, income statement, and cash flow statement. By analyzing the three major financial statements (balance sheet, income statement, and cash flow statement) according to conventional financial formulas, and combining national standards and the characteristics of the target company, the corresponding asset and liability dataset for the target company is set.

[0076] Specifically, first, determine the target company's industry, development stage, strategic planning, and other corporate background information; then, based on the Accounting Standards for Business Enterprises and commonly used financial formulas, correctly handle matters such as fixed asset depreciation and R&D expenditure capitalization to determine the relationship between assets, liabilities, and owner's equity, and determine business logic data such as current ratio, quick ratio, and debt-to-equity ratio; finally, through the above-mentioned layered processing and data sorting, set up the corresponding asset and liability dataset for the target company.

[0077] In this embodiment of the application, based on the centralized data information of the asset and liability dataset, and combined with national standard requirements and benchmarking data related to the asset and liability of listed companies in the industry, a comprehensive review and data analysis of the asset and liability indicator system is conducted. The data information is extracted and summarized to construct the asset and liability indicator system corresponding to the target enterprise.

[0078] Here, the indicator system includes multiple indicator dimensions and at least one financial indicator corresponding to each indicator dimension, as well as the indicator calculation formula and data source corresponding to each financial indicator; wherein, the indicator dimension includes multiple first indicator dimensions and at least one second indicator dimension under each first indicator dimension.

[0079] In one possible implementation of this application, the first indicator dimension may include, but is not limited to, net working capital, fixed assets, debt, equity, and cash; wherein, the second indicator dimension corresponding to the net working capital may include, but is not limited to, sales revenue, cost of sales, cost of goods sold, and operating efficiency; the second indicator dimension corresponding to the fixed assets may include, but is not limited to, tangible assets, intangible assets, and goodwill; the second indicator dimension corresponding to the debt may include, but is not limited to, debt at the beginning of the year and changes in debt; and the second indicator dimensions corresponding to the equity and cash are equity and cash itself, respectively.

[0080] Thus, the aforementioned indicator system is a multi-level financial indicator system design. This structure helps to more comprehensively and meticulously evaluate a company's financial condition and operating performance. Specifically, this indicator system has the following technical effects.

[0081] The systematic and structured approach, by dividing financial indicators into multiple dimensions (first indicator dimension) and further subdividing each dimension into second indicator dimensions, makes the entire evaluation system more systematic and structured, which helps to comprehensively analyze the financial health of enterprises from different perspectives.

[0082] Refined management, with at least one secondary indicator dimension under each primary indicator dimension, allows for a more in-depth examination of specific aspects; for example, under the primary indicator dimension of "profitability," there might be specific secondary indicator dimensions such as net profit margin and gross profit margin. This helps companies better understand their performance in various aspects and take targeted improvement measures.

[0083] A comprehensive evaluation capability and a multi-dimensional indicator system allow companies to examine their performance from multiple perspectives, such as financial stability, operational efficiency, and market competitiveness. This is not only beneficial for internal management decisions but also helps external investors or creditors gain a comprehensive understanding of the company.

[0084] It is highly adaptable. Because the system is built on multiple levels, it can flexibly adjust the focus of attention according to different industry characteristics or enterprise development stages. For example, for start-ups, more emphasis may be placed on cash flow-related indicators, while for mature enterprises, more emphasis may be placed on profit growth and return on assets.

[0085] Improving transparency by clearly listing various financial indicators and their corresponding dimensions enhances information transparency and plays a positive role in boosting stakeholder confidence and promoting the healthy development of the capital market.

[0086] Supporting strategic planning and execution, this detailed financial data allows management to more accurately formulate long-term development strategies and assess the effectiveness of strategy implementation by regularly monitoring changes in key indicators, enabling timely adjustments to strategies to respond to market changes.

[0087] Risk control, through continuous monitoring of various financial indicators, can help companies identify potential problem areas early, such as working capital shortages and poor cost control, thereby taking preventative measures to reduce financial risks.

[0088] In one embodiment of this application, in specific implementation, the step S101 of constructing the asset and liability indicator system corresponding to the target enterprise based on the preset asset and liability dataset corresponding to the target enterprise may include:

[0089] S1011. Based on the preset asset and liability dataset corresponding to the target enterprise, and combined with the preset accounting dimensions and project data, determine the basic financial data corresponding to the asset and liability dataset.

[0090] In this step, based on the account transaction amounts and cumulative balance data obtained from the financial system in the asset and liability dataset corresponding to the target enterprise, combined with the preset accounting dimensions and the basic data of the planned project, the basic financial data corresponding to the asset and liability dataset is determined.

[0091] The basic financial data may include, but is not limited to, operating revenue, fixed assets, intangible assets, operating cash flow, investing cash flow, current liabilities, non-current liabilities, financial expenses, cash and cash equivalents, and financing gap.

[0092] The accounting dimensions refer to different aspects or perspectives used to classify and organize financial data. These dimensions help enterprises to be more systematic and structured in recording, reporting, and analyzing financial information. For example, time dimension, geographical dimension, product and service dimension, customer dimension, sales dimension, and accounting dimension. Through these accounting dimensions, enterprises can analyze their financial situation more precisely, identify problems, and formulate corresponding improvement measures. At the same time, these dimensions also support the management's need for financial information in the decision-making process, improving the efficiency and effectiveness of financial management.

[0093] The project data refers to the basic data determined through project planning. That is, a series of key information that needs to be collected and prepared before the project starts or during the project execution. This data is crucial for the planning, execution, monitoring and ultimate success of the project. For example, project background and objectives, project scope, time plan, resource requirements, risk management and quality management, etc. Project data provides solid support for the smooth progress of the project, ensuring that the project team can clearly understand the direction, objectives and requirements of the project, thereby improving the success rate of the project.

[0094] In this way, by combining the preset accounting dimensions and project data, the basic financial data corresponding to the asset and liability dataset is determined, realizing refined management of the target enterprise in various accounting dimensions. It can also support cross-analysis by combining multiple accounting dimensions (such as time, region, product, etc.), and can comprehensively assess the financial health of the enterprise from multiple perspectives, discover potential problems and opportunities. In addition, through detailed project data, the enterprise can conduct comprehensive financial analysis, including profitability, solvency, operational efficiency and other aspects, so as to formulate more scientific and reasonable strategic plans.

[0095] S1012. Perform data cleaning processing on the basic financial data to obtain the indicator system data corresponding to the basic financial data, and determine the mapping relationship of the basic financial data.

[0096] In this step, based on the pre-determined system standards of the asset and liability indicator system corresponding to the target enterprise, the basic financial data is cleaned and processed to obtain the indicator system data corresponding to the basic financial data, and the mapping relationship of the basic financial data is determined to unify the scope of the basic financial data and improve the integrity and usability of the data.

[0097] The mapping relationship of the basic financial data refers to the correlation between various basic financial data, such as the investment relationship between current liabilities, non-current liabilities and investment cash flow.

[0098] S1013. Based on the indicator system data and the mapping relationship, and combined with the preset debt risk quantitative assessment system and the operating characteristics of the target enterprise, construct the asset and liability indicator system corresponding to the target enterprise.

[0099] In this step, by comparing the data with the pre-set debt risk quantitative assessment system and combining it with the operating characteristics of the target company, the data of the indicator system is analyzed in a data-driven manner, and the data is classified and summarized according to the mapping relationship and financial data attributes to construct the asset and liability indicator system corresponding to the target company.

[0100] The target company's business characteristics include information such as industry characteristics, products and services, market positioning and competition, business model, financial status, organizational management structure, and risk management and strategic planning.

[0101] Here, in the asset and liability indicator system, the data sources corresponding to each financial indicator may include, but are not limited to, financial statements such as the balance sheet, income statement, and cash flow statement. Since the financial statements interact with each other and are related, the first indicator dimension may include net working capital, fixed assets, liabilities, equity, and cash to ensure full coverage of the mutual influence of each indicator dimension.

[0102] For example, the following table shows an example of an asset and liability indicator system for a target company.

[0103]

[0104]

[0105] S102. Based on the asset and liability indicator system, using the preset multiple prediction method and combined with external adjustment indication information, the predicted value of each financial indicator at a preset future time point is obtained.

[0106] In this embodiment of the application, the preset future time point can be specifically set according to actual needs and the financial plan of the target enterprise.

[0107] In one embodiment of this application, step S102 may include:

[0108] S1021. Based on the data source corresponding to each of the financial indicators, the actual value of the financial indicator at the current time point is obtained using the indicator calculation formula.

[0109] In this step, based on the data source corresponding to each financial indicator, each financial indicator is input into the corresponding indicator calculation formula, and the actual value of the financial indicator at the current time point is obtained through the formula calculation.

[0110] S1022. Based on the actual value, the predicted multiple value corresponding to the financial indicator is obtained using a preset multiple prediction method.

[0111] In this step, in practice, firstly, based on the target company's financial situation and operating characteristics, the predicted multiple corresponding to each financial indicator is determined; then, based on the predicted multiple, the actual value is predicted by multiple; finally, the predicted multiple value corresponding to the financial indicator is obtained.

[0112] For example, when determining the forecast multiple for revenue growth rate, it is necessary to refer to the average growth rate of other companies in the same industry, analyze the company's revenue growth rate over the past few years, make adjustments based on the current market environment and the company's development stage, and finally assess the growth potential of the target market to determine the forecast multiple.

[0113] For example, when forecasting the debt-to-equity ratio, one can refer to the average debt-to-equity ratio of the same industry, analyze the company's debt-to-equity ratio over the past few years, consider changes in financing strategies and capital structure, and finally adjust the forecast multiple if the company has a clear financing plan (such as issuing bonds or issuing additional shares).

[0114] In addition, the forecast multiple corresponding to a financial indicator can be determined based on the historical growth experience of the financial indicator.

[0115] S1023. In response to receiving an external adjustment instruction to adjust the multiple prediction value, the multiple prediction value is adjusted based on the external adjustment instruction and the future time point corresponding to the preset prediction period to obtain the prediction value of the financial indicator at the future time point.

[0116] In this step, in specific implementation, firstly, the system receives external adjustment instructions from the user to adjust the multiple forecast value; then, in response to the external adjustment instructions, the system determines the content of the external adjustment instructions and the future time point corresponding to the preset forecast period; finally, based on the content of the external adjustment instructions, the multiple forecast value is adjusted to obtain the forecast value of the financial indicator at the future time point.

[0117] For example, please refer to Figure 2 , Figure 2 This is a chart illustrating a predicted financial indicator of net working capital, provided as an embodiment of this application. Figure 2 As shown, in the financial indicators under net working capital, in addition to current assets such as cash, accounts receivable, and inventory, and common current liabilities such as accounts payable and short-term loans, there may also be other current assets or liabilities, such as prepayments and advance payments. Since the amount and changes of other current assets or liabilities are more complex, and there is no clear reason to indicate that they will be affected by other factors, other current assets or liabilities are estimated as a percentage weighted average of sales revenue or cost of goods sold to predict the forecast value of other current assets or liabilities; wherein, the percentage can be specifically marked according to the actual financial situation of the target company.

[0118] Furthermore, based on the projected values ​​of each financial indicator at future points in time, the relationship between the annual changes of each indicator dimension and the cash flow statement can be calculated.

[0119] For example, please refer to Figure 3 , Figure 3 This is a schematic diagram illustrating the relationship between net working capital and the cash flow statement, provided as an embodiment of this application. Figure 3 As shown, if net working capital increases, cash flow will decrease due to an outflow; if net working capital decreases, cash flow will increase due to an inflow.

[0120] Furthermore, regarding fixed assets in the first indicator dimension, since capital expenditure refers to increased investment in fixed assets, but as assets age, depreciation or amortization must be deducted. To simplify the estimation of depreciation and amortization, fixed assets are linked to sales revenue when managing the asset and liability indicator system. For example, since goodwill is a one-off event and often occurs in mergers and acquisitions, its value is usually kept unchanged and not impaired. Thus, when consolidating the cash flow statement, cash paid for the construction of fixed assets, intangible assets, and other long-term assets will affect the cash flows generated by investing activities in the cash flow statement.

[0121] Furthermore, regarding debt in the first indicator dimension, when forecasting debt, it is necessary to pay attention to debt changes and accumulate the debt at the beginning of the year. Unless otherwise specified, it is assumed that no new borrowing will be made in the future. Since the repayment time of the debt to be repaid in the future can be found in the terms of the loan agreement, the forecast of debt changes is relatively controllable. Here, the debt change includes not only the principal to be repaid but also the interest to be paid this year. The interest rate can also be found in the terms of the loan agreement. Thus, when consolidating the cash flow statement, the cash outflows generated by debt repayment and interest payment affect the cash flow generated by financing activities in the cash flow statement.

[0122] Regarding equity in the first indicator dimension, when forecasting equity on the balance sheet, it is necessary to pay attention to changes in equity and accumulate the beginning value and net profit, and then subtract dividends; in this way, when consolidating the cash flow statement, the impact of equity on cash flow is reflected in the cash flow generated by financing activities in the cash flow statement.

[0123] Furthermore, regarding cash in the first indicator dimension, by calculating the annual change of cash in the cash flow statement, the effects of the above indicator dimensions are reflected in each indicator in the cash flow statement. Then, by adding the previous year's cash balance to it, the year-end cash balance can be predicted.

[0124] S103. Based on the predicted values, a pre-trained multiple linear regression model is used to predict the target company's predicted debt-to-asset ratio at the future time point.

[0125] It should be noted that the task of the aforementioned multiple linear regression model is to analyze the impact of multiple independent variables in financial data on the dependent variable, in order to more comprehensively explain the changes in the dependent variable.

[0126] The independent variables include financial indicators such as cash, assets, and liabilities; the dependent variable is the predicted debt-to-asset ratio, and the financial indicators are identified as the sensitive factors in the multiple linear regression model.

[0127] Furthermore, sensitive factors are extracted through feature engineering, and all feature attributes are standardized through data cleaning, outlier handling, and normalization to improve model accuracy. In particular, the normalization of sensitive factors addresses the problem of excessive dimensional differences and avoids situations where the update process does not always move towards the minimum point. Essentially, it transforms dimensional data into dimensionless data and converts the data to the same order of magnitude, thus solving the problem of incomparability due to large differences in the order of magnitude between data.

[0128] In one embodiment of this application, the step of pre-training the multiple linear regression model in step S103 may include:

[0129] Step A: Determine the financial indicators as the independent variables of the multiple linear regression model, and determine the debt-to-equity ratio as the dependent variable of the multiple linear regression model.

[0130] Step B: Based on the asset and liability dataset and the asset and liability indicator system, determine the corresponding linear relationship between the independent variable and the dependent variable, and obtain the calculation expression of the multiple linear regression model to obtain the multiple linear regression model to be trained.

[0131] Step C: Use the preset residual method to check whether the multiple linear regression model to be trained satisfies the assumptions corresponding to the multiple linear regression model.

[0132] In this step, the preset residual method may include, but is not limited to, the checking methods corresponding to residual independence, homoscedasticity, and normal distribution.

[0133] Step D: If satisfied, perform feature selection and feature construction on the independent variable, and determine the direction of influence of the independent variable.

[0134] Step E: Based on the asset and liability dataset and the direction of influence, train the multiple linear regression model to be trained using a preset training method, and optimize the performance of the multiple linear regression model to be trained to obtain the weight parameters corresponding to the calculation expression, so as to obtain the trained multiple linear regression model.

[0135] In this step, the preset training method can be an algorithm such as least squares; the mean square error is used to evaluate the model's performance metrics, and the model's performance is optimized through a rule engine and by debugging the model's hyperparameters.

[0136] In this embodiment of the application, since it is difficult to identify and label sensitive factors affecting capital structure and business operations, this embodiment addresses the above problem by clarifying business objectives, defining the life cycle of asset and liability classifications and hierarchical governance of indicators, and constructing an asset and liability indicator system. By clarifying objectives and asset and liability classifications and combining expert experience, sensitive factors with large predictive biases in the asset and liability indicator system are identified, effectively solving the problem of sensitive factor labeling. By constructing an asset and liability indicator system and training the model, comprehensive and systematic financial data information can be obtained efficiently, ensuring the accuracy and practicality of model training.

[0137] Furthermore, for financial indicators with complex business logic in the asset-liability indicator system, multidimensional tables can be designed to modularize and granulate the complex business logic based on the listed different scenarios or combinations of scenarios. This reduces the complexity of the multiple linear regression model during simulation and reasoning, as well as the probability of failing to identify differences due to similar scenarios, thereby improving the accuracy of the multiple linear regression model in predicting the asset-liability ratio.

[0138] Furthermore, to address the issue of frequent changes in sensitive factors, this application embodiment can employ a flexible sensitive factor update mechanism. When changes in business logic cause changes in sensitive factors, the update mechanism can simplify the additional development workload caused by these changes, while eliminating the need to retrain the multiple linear regression model, thereby improving the flexibility and stability of the multiple linear regression model.

[0139] In one embodiment of this application, step S103 may include:

[0140] S1031. Based on the weight parameters and the operating conditions of the target enterprise, the actual weight parameters are determined through a preset scientific calculation model.

[0141] In this step, the weight parameters output by the multiple linear regression model through model training are adjusted based on the experience of business experts and relevant scientific computing models to obtain the actual weight parameters.

[0142] S1032. Based on the predicted values ​​and the actual weight parameters, the predicted debt-to-asset ratio of the target enterprise at the future time point is obtained using the calculation expression in the multiple linear regression model.

[0143] In this step, the predicted value and actual weight parameters corresponding to each financial indicator are input into the calculation expression in the multiple linear regression model, and the predicted debt-to-equity ratio of the target company at future time points is obtained through calculation.

[0144] Specifically, the calculation expression is as follows.

[0145] Y = β0 + β1X1 + β2X2 + ... + β n X n +∈.

[0146] Where Y represents the projected debt-to-equity ratio of the target company at future points in time; X1, X2, ..., X n These are the independent variables (financial indicators); β0, β1, β2, ..., β n These are the actual weight parameters; ∈ represents the error term (the variation that the model fails to explain).

[0147] S104. Compare the predicted debt-to-asset ratio with the pre-configured early warning threshold to obtain the comparison result, and issue an early warning regarding the debt-to-asset ratio to the target enterprise based on the comparison result.

[0148] In this embodiment of the application, a risk rule management model for the target enterprise is constructed using a rule engine based on financial indicators in the asset and liability indicator system, namely, sensitive factors in the multiple linear regression model; wherein, the risk rule management model includes pre-configured early warning thresholds that combine business expert experience and relevant scientific computing models.

[0149] Here, the risk rule management model determines the order of forecasting financial indicators. Specifically, first, operating revenue and corresponding financial expenses are forecasted in the income statement; then, other asset and liability related items are calculated based on the financial forecast assumptions in the income statement; finally, based on the forecast results of the income statement and balance sheet, the indirect method is used to forecast the financial indicators in the cash flow statement.

[0150] In this step, when the predicted debt-to-asset ratio is less than or equal to the pre-configured warning threshold, the warning for the target company regarding the debt-to-asset ratio is only manifested as visual monitoring; when the predicted debt-to-asset ratio is greater than the pre-configured warning threshold, the target company is given a warning and reminder regarding the debt-to-asset ratio.

[0151] Optional, please refer to Figure 4 , Figure 4 This is a second flowchart illustrating a method for early warning of debt-to-equity ratio provided in an embodiment of this application. Figure 4 As shown, in addition to the early warning method for the debt-to-asset ratio described in steps S101 to S104, step S105 is also included. Specifically, step S105 is used to explain the method of visually displaying the relevant data of the debt-to-asset ratio based on the changes in the debt-to-asset ratio corresponding to the time line, so as to improve the user's experience when doing financial work.

[0152] S105. Based on the predicted debt-to-asset ratio and the historical debt-to-asset ratio of the target enterprise obtained from the preset historical dataset, determine the backtest value of the debt-to-asset ratio of the target enterprise, and visualize the predicted debt-to-asset ratio, the historical debt-to-asset ratio, and the backtest value of the debt-to-asset ratio to the outside world.

[0153] In this step, the specific implementation involves first obtaining the historical debt-to-asset ratio of the target company from a preset historical dataset; then, based on the predicted debt-to-asset ratio and the historical debt-to-asset ratio, calculating the backtest value of the target company's corresponding debt-to-asset ratio; and finally, visualizing the predicted debt-to-asset ratio, the historical debt-to-asset ratio, and the backtest value of the debt-to-asset ratio to the outside world through visualization processing.

[0154] In summary, the embodiments of this application aim to achieve classified management of the debt-to-asset ratio of enterprises in the energy industry, maintain a reasonable debt level, and promote the rapid return of high debt to a reasonable level. By accurately predicting the debt-to-asset ratio and monitoring the enterprise's debt-to-asset ratio and capital structure in real time, intelligent prediction services are provided for sensitive factors in the energy industry that change frequently and are difficult to predict manually. Furthermore, the prediction of the enterprise's capital structure is achieved through models such as deep learning and machine learning.

[0155] The asset-liability ratio early warning method provided in this application constructs an asset-liability indicator system for the target enterprise and determines the financial indicators under this system. It predicts the value of each financial indicator at a future time point, and then uses a multiple linear regression model to predict the asset-liability ratio. By comparing the predicted asset-liability ratio with the early warning threshold, it achieves early warning of the asset-liability ratio of the target enterprise. This method realizes the effect of predicting the asset-liability ratio based on comprehensive data and financial indicators, and improves the accuracy and efficiency of early warning of the enterprise's asset-liability ratio.

[0156] Please see Figure 5 , Figure 6 , Figure 5 This is one of the structural schematic diagrams of an early warning device for the debt-to-equity ratio provided in an embodiment of this application. Figure 6 This is a second schematic diagram of a debt-to-equity ratio early warning device provided in an embodiment of this application. Figure 5 As shown, the early warning device 500 includes:

[0157] The indicator system module 510 is used to construct an asset and liability indicator system for the target enterprise based on a preset asset and liability dataset corresponding to the target enterprise; wherein, the indicator system includes multiple indicator dimensions and at least one financial indicator corresponding to each indicator dimension, as well as the indicator calculation formula and data source for each financial indicator;

[0158] The first prediction module 520 is used to obtain the predicted value of each financial indicator at a preset future time point based on the asset and liability indicator system, using a preset multiple prediction method and combined with external adjustment indication information.

[0159] The second prediction module 530 is used to make a prediction based on the predicted value using a pre-trained multiple linear regression model to obtain the predicted debt-to-asset ratio of the target enterprise at the future time point.

[0160] The comparison and early warning module 540 is used to compare the predicted debt-to-asset ratio with a pre-configured early warning threshold, obtain the comparison result, and issue an early warning to the target enterprise regarding the debt-to-asset ratio based on the comparison result.

[0161] Furthermore, when the indicator system module 510 is used to construct the asset and liability indicator system corresponding to the target enterprise based on the preset asset and liability dataset corresponding to the target enterprise, the indicator system module 510 is used for:

[0162] Based on the preset asset and liability dataset corresponding to the target enterprise, and combined with the preset accounting dimensions and project data, the basic financial data corresponding to the asset and liability dataset is determined.

[0163] The basic financial data is cleaned to obtain the indicator system data corresponding to the basic financial data, and the mapping relationship of the basic financial data is determined.

[0164] Based on the data of the indicator system and the mapping relationship, combined with the preset debt risk quantitative assessment system and the operating characteristics of the target enterprise, an asset and liability indicator system corresponding to the target enterprise is constructed.

[0165] Furthermore, when the first prediction module 520 is used to obtain the predicted value of each financial indicator at a preset future time point based on the asset-liability indicator system, using a preset multiple prediction method and combining external adjustment indication information, the first prediction module 520 is used to:

[0166] Based on the data source corresponding to each of the financial indicators, the actual value of the financial indicator at the current point in time is obtained using the indicator calculation formula.

[0167] Based on the actual values, the predicted multiple value corresponding to the financial indicator is obtained using a preset multiple prediction method.

[0168] In response to receiving an external adjustment instruction to adjust the predicted multiple value, the predicted multiple value is adjusted based on the external adjustment instruction and the future time point corresponding to the preset prediction period, so as to obtain the predicted value of the financial indicator at the future time point.

[0169] Furthermore, when the second prediction module 530 is used to pre-train the multiple linear regression model, the second prediction module 530 is used to:

[0170] The financial indicators are determined as the independent variables of the multiple linear regression model, and the debt-to-equity ratio is determined as the dependent variable of the multiple linear regression model.

[0171] Based on the asset and liability dataset and the asset and liability indicator system, the corresponding linear relationship between the independent variable and the dependent variable is determined, and the calculation expression of the multiple linear regression model is obtained to obtain the multiple linear regression model to be trained.

[0172] The preset residual method is used to check whether the multivariate linear regression model to be trained satisfies the assumptions corresponding to the multivariate linear regression model.

[0173] If the conditions are met, feature selection and feature construction are performed on the independent variable, and the direction of influence of the independent variable is determined.

[0174] Based on the asset and liability dataset and the direction of influence, the multiple linear regression model to be trained is trained using a preset training method, and the performance of the multiple linear regression model to be trained is optimized to obtain the weight parameters corresponding to the calculation expression, so as to obtain the trained multiple linear regression model.

[0175] Furthermore, when the second prediction module 530 is used to predict the target company's debt-to-equity ratio at the future time point based on the predicted value using a pre-trained multiple linear regression model, the second prediction module 530 is used to:

[0176] Based on the weight parameters and the operating conditions of the target enterprise, the actual weight parameters are determined through a preset scientific calculation model.

[0177] Based on the predicted values ​​and the actual weight parameters, the predicted debt-to-asset ratio of the target enterprise at the future time point is obtained using the calculation expression in the multiple linear regression model.

[0178] Furthermore, such as Figure 6 As shown, the early warning device 500 further includes a data display module 550, which is used for:

[0179] Based on the predicted debt-to-asset ratio and the historical debt-to-asset ratio of the target company obtained from the preset historical dataset, the backtest value of the debt-to-asset ratio of the target company is determined, and the predicted debt-to-asset ratio, the historical debt-to-asset ratio, and the backtest value of the debt-to-asset ratio are visualized externally.

[0180] The asset-liability ratio early warning device provided in this application constructs an asset-liability indicator system for the target enterprise and determines the financial indicators under this system. It predicts the predicted value of each financial indicator at a future time point, and then uses a multiple linear regression model to predict the asset-liability ratio. By comparing the predicted asset-liability ratio with the early warning threshold, it can provide early warning about the asset-liability ratio of the target enterprise. This achieves the effect of predicting the asset-liability ratio based on comprehensive data and financial indicators, and improves the accuracy and efficiency of early warning of the enterprise's asset-liability ratio.

[0181] Please see Figure 7 , Figure 7 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Figure 7 As shown, the electronic device 700 includes a processor 710, a memory 720, and a bus 730.

[0182] The memory 720 stores machine-readable instructions executable by the processor 710. When the electronic device 700 is running, the processor 710 communicates with the memory 720 via the bus 730. When the machine-readable instructions are executed by the processor 710, they can perform the operations described above. Figure 1 as well as Figure 2 The steps of the asset-liability ratio early warning method in the method embodiment shown are described in detail in the method embodiment, and will not be repeated here.

[0183] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, can perform the above-described actions. Figure 1 as well as Figure 2 The steps of the asset-liability ratio early warning method in the method embodiment shown are described in detail in the method embodiment, and will not be repeated here.

[0184] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0185] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the shown or discussed mutual couplings, direct couplings, or communication connections may be through some communication interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.

[0186] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0187] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0188] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a processor-executable, non-volatile, computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0189] Finally, it should be noted that the above-described embodiments are merely specific implementations of this application, used to illustrate the technical solutions of this application, and not to limit them. The scope of protection of this application is not limited thereto. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments, or make equivalent substitutions for some of the technical features, within the scope of the technology disclosed in this application. Such modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for early warning of debt-to-asset ratio, characterized in that, The early warning method includes: Based on a pre-defined asset and liability dataset corresponding to the target enterprise, an asset and liability indicator system corresponding to the target enterprise is constructed; wherein, the indicator system includes multiple indicator dimensions and at least one financial indicator corresponding to each indicator dimension, as well as the indicator calculation formula and data source for each financial indicator; Based on the aforementioned asset and liability indicator system, using a preset multiple prediction method and combined with external adjustment indication information, the predicted value of each financial indicator at a preset future time point is obtained. Based on the predicted values, a pre-trained multiple linear regression model is used to predict the target company's projected debt-to-equity ratio at the future time point. The predicted debt-to-asset ratio is compared with a pre-configured early warning threshold to obtain a comparison result. Based on the comparison result, an early warning is issued to the target company regarding its debt-to-asset ratio.

2. The method according to claim 1, characterized in that, The construction of the asset and liability indicator system for the target enterprise based on the pre-set asset and liability dataset includes: Based on the preset asset and liability dataset corresponding to the target enterprise, and combined with the preset accounting dimensions and project data, the basic financial data corresponding to the asset and liability dataset is determined. The basic financial data is cleaned to obtain the indicator system data corresponding to the basic financial data, and the mapping relationship of the basic financial data is determined. Based on the data of the indicator system and the mapping relationship, combined with the preset debt risk quantitative assessment system and the operating characteristics of the target enterprise, an asset and liability indicator system corresponding to the target enterprise is constructed.

3. The method according to claim 1, characterized in that, Based on the aforementioned asset-liability indicator system, the method utilizes a preset multiple forecasting method combined with external adjustment indications to obtain the predicted value of each financial indicator at a preset future time point, including: Based on the data source corresponding to each of the financial indicators, the actual value of the financial indicator at the current point in time is obtained using the indicator calculation formula. Based on the actual values, the predicted multiple value corresponding to the financial indicator is obtained using a preset multiple prediction method. In response to receiving an external adjustment instruction to adjust the predicted multiple value, the predicted multiple value is adjusted based on the external adjustment instruction and the future time point corresponding to the preset prediction period, so as to obtain the predicted value of the financial indicator at the future time point.

4. The method according to claim 1, characterized in that, The multiple linear regression model is pre-trained using the following steps: The financial indicators are determined as the independent variables of the multiple linear regression model, and the debt-to-equity ratio is determined as the dependent variable of the multiple linear regression model. Based on the asset and liability dataset and the asset and liability indicator system, the corresponding linear relationship between the independent variable and the dependent variable is determined, and the calculation expression of the multiple linear regression model is obtained to obtain the multiple linear regression model to be trained. The preset residual method is used to check whether the multivariate linear regression model to be trained satisfies the assumptions corresponding to the multivariate linear regression model. If the conditions are met, feature selection and feature construction are performed on the independent variable, and the direction of influence of the independent variable is determined. Based on the asset and liability dataset and the direction of influence, the multiple linear regression model to be trained is trained using a preset training method, and the performance of the multiple linear regression model to be trained is optimized to obtain the weight parameters corresponding to the calculation expression, so as to obtain the trained multiple linear regression model.

5. The method according to claim 4, characterized in that, The step of using a pre-trained multiple linear regression model to predict the target company's projected debt-to-equity ratio at the future time point, based on the predicted values, includes: Based on the weight parameters and the operating conditions of the target enterprise, the actual weight parameters are determined through a preset scientific calculation model. Based on the predicted values ​​and the actual weight parameters, the predicted debt-to-asset ratio of the target enterprise at the future time point is obtained using the calculation expression in the multiple linear regression model.

6. The method according to claim 1, characterized in that, Before comparing the debt-to-equity ratio with a pre-configured warning threshold, the method further includes: Based on the predicted debt-to-asset ratio and the historical debt-to-asset ratio of the target company obtained from the preset historical dataset, the backtest value of the debt-to-asset ratio of the target company is determined, and the predicted debt-to-asset ratio, the historical debt-to-asset ratio, and the backtest value of the debt-to-asset ratio are visualized externally.

7. A debt-to-asset ratio early warning device, characterized in that, The early warning device includes: The indicator system module is used to construct an asset and liability indicator system for target companies in the energy industry based on a preset asset and liability dataset for the target companies. The indicator system includes multiple indicator dimensions and at least one financial indicator for each indicator dimension, as well as the indicator calculation formula and data source for each financial indicator. The first prediction module is used to obtain the predicted value of each financial indicator at a preset future time point based on the asset and liability indicator system, using a preset multiple prediction method and combined with external adjustment indication information. The second prediction module is used to make predictions based on the predicted values ​​using a pre-trained multiple linear regression model, and to obtain the predicted debt-to-asset ratio of the target enterprise at the future time point. The comparison and early warning module is used to compare the predicted debt-to-asset ratio with a pre-configured early warning threshold, obtain the comparison result, and issue an early warning to the target company regarding the debt-to-asset ratio based on the comparison result.

8. The apparatus according to claim 7, characterized in that, The early warning device also includes a data display module, which is used for: Based on the predicted debt-to-asset ratio and the historical debt-to-asset ratio of the target company obtained from the preset historical dataset, the backtest value of the debt-to-asset ratio of the target company is determined, and the predicted debt-to-asset ratio, the historical debt-to-asset ratio, and the backtest value of the debt-to-asset ratio are visualized externally.

9. An electronic device, characterized in that, include: The device includes a processor, a memory, and a bus. The memory stores machine-readable instructions executable by the processor. When the electronic device is running, the processor communicates with the memory via the bus. The machine-readable instructions are executed by the processor to perform the steps of the debt-to-equity ratio early warning method as described in any one of claims 1 to 6.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, performs the steps of the debt-to-equity ratio early warning method as described in any one of claims 1 to 6.

Images