Reserved resource service data analysis method, apparatus and device, and medium

Abstract

The invention relates to the technical field of data processing and the field of financial science and technology, and discloses a reserve resource business data analysis method, which comprises the following steps: collecting reserve resource business data in a preset time range, carrying out statistical analysis, determining a target analysis model, and optimizing and updating parameters of the target analysis model based on historical business data. And processing the reserve resource service data to generate a preliminary analysis result, adjusting the preliminary analysis result by applying an adaptive adjustment mode, and generating an analysis result under the multi-service scene. According to the method, the model is determined based on data statistical analysis, and the model parameters are optimized and updated, so that the accuracy and adaptability of the model are improved. Flexible adjustment of multiple service scenes is realized through an adaptive adjustment mode, the risk of mobility of reserve resources under different market conditions is effectively dealt with, and accurate analysis and dynamic management of reserve resource data are realized.

Description

Technical Field

[0001] The present invention relates to the fields of data processing technology and financial technology, and in particular to a reserve resource business data analysis method, device, equipment and storage medium. Background Art

[0002] In the financial sector, the management of reserve resources (such as provident funds) is crucial for banks and other financial institutions. As part of institutional strategic customers, provident funds are an important source of low-cost liabilities for financial institutions. At the same time, they have high value and high retention rates when serving C-end customers.

[0003] In order to ensure the stability of the financial system, the Provident Fund Management Center currently relies mainly on external and internal compliance standards for liquidity management.

[0004] For internal compliance management, banks usually refer to the internal monitoring indicator system. These internal indicator systems include but are not limited to liquidity risk measurement management methods, liquidity risk management procedures, liquidity risk early warning mechanism and emergency plan management procedures. The key indicators for specific monitoring include deposit-loan ratio, core deposit ratio, deposit concentration, liquid asset ratio and excess reserve ratio.

[0005] Although the existing compliance framework can guide the Provident Fund Management Center to manage liquidity to a certain extent, due to the lack of experience in bank fund management, the Provident Fund Management Center usually only relies on historical experience to adjust the liquidity and risk factors of the Provident Fund funds. This adjustment method is difficult to effectively predict and respond to the liquidity risk of funds when there are fluctuations in the real estate market, resulting in partial bank runs and liquidity risks when the market fluctuates.

[0006] The shortcomings of existing technologies in addressing provident fund management are mainly manifested in the lack of scientific and effective methods for predicting and dynamically adjusting fund liquidity, and the inability to adapt to external market fluctuations and changes in different business scenarios. When faced with different economic environments and market changes, the provident fund management center relies on manual experience to adjust funds, resulting in low accuracy in liquidity risk prediction and control, and a large hidden danger of fund risk. Summary of the invention

[0007] The main purpose of the present invention is to provide a reserve resource business data analysis method, device, equipment and storage medium, aiming to solve the technical problem that the prior art lacks scientific analysis models and adaptive adjustment means, making it difficult to effectively predict and dynamically adjust the liquidity of reserve resources to cope with liquidity risks brought about by different business scenarios and market changes.

[0008] To achieve the above object, the present invention provides a method for analyzing reserve resource business data, comprising:

[0009] Collect reserve resource business data within a preset time range;

[0010] Performing statistical analysis on the reserve resource business data, and determining a target analysis model based on the statistical analysis results;

[0011] Based on the historical business data of the reserve resources, the parameters of the target analysis model are optimized and updated;

[0012] Using the target analysis model to process the reserve resource business data within the preset time range to generate a preliminary analysis result;

[0013] The preliminary analysis results are adjusted through an adaptive adjustment mode to generate analysis results in multiple business scenarios.

[0014] Furthermore, to achieve the above-mentioned purpose, the present invention provides a reserve resource business data analysis device, comprising:

[0015] A data collection module is used to collect the reserve resource business data within a preset time range;

[0016] A data analysis and model selection module, used to perform statistical analysis on the reserve resource business data and determine a target analysis model based on the statistical analysis results;

[0017] A model optimization and training module, used to optimize and update the parameters of the target analysis model based on the historical business data of the reserve resources;

[0018] A model application and analysis module, used to use the target analysis model to process the reserve resource business data within the preset time range to generate preliminary analysis results;

[0019] The result adjustment and scenario management module is used to adjust the preliminary analysis results through an adaptive adjustment mode to generate analysis results in multiple business scenarios.

[0020] Furthermore, to achieve the above-mentioned purpose, the present invention also provides a computer device, which includes a memory, a processor, and a reserve resource business data analysis program stored in the memory and executable on the processor, and the reserve resource business data analysis program, when executed by the processor, implements the steps of the reserve resource business data analysis method as described above.

[0021] Furthermore, to achieve the above-mentioned purpose, the present invention also provides a computer-readable storage medium, on which a reserve resource business data analysis program is stored, and when the reserve resource business data analysis program is executed by a processor, the steps of the reserve resource business data analysis method as described above are implemented.

[0022] Beneficial effects: The present invention relates to the fields of data processing technology and financial technology, and discloses a method for analyzing reserve resource business data, which collects reserve resource business data within a preset time range, performs statistical analysis and determines a target analysis model, optimizes and updates the parameters of the target analysis model based on historical business data, processes the reserve resource business data to generate preliminary analysis results, and applies an adaptive adjustment mode to adjust the preliminary analysis results to generate analysis results in multiple business scenarios. The present invention improves the accuracy and adaptability of the model by determining a model based on data statistical analysis and optimizing and updating the model parameters. Flexible adjustment of multiple business scenarios is achieved through the adaptive adjustment mode, which effectively responds to the liquidity risks of reserve resources under different market conditions, and realizes accurate analysis and dynamic management of reserve resource data. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The present invention will be further described below with reference to the accompanying drawings and embodiments, in which:

[0024] Figure 1 A schematic diagram of an application environment of a method for analyzing reserve resource business data in an embodiment of the present invention;

[0025] Figure 2 A schematic diagram of a flow chart of an embodiment of a method for analyzing data of a reserve resource business according to the present invention;

[0026] Figure 3 A schematic diagram of functional modules of a preferred embodiment of the reserve resource business data analysis device of the present invention;

[0027] Figure 4 A schematic diagram of the structure of a computer device in one embodiment of the present invention;

[0028] Figure 5 FIG. 4 is another schematic diagram of the structure of a computer device in one embodiment of the present invention. DETAILED DESCRIPTION

[0029] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0030] The reserve resource service data analysis method provided by the embodiment of the present invention can be applied in the following aspects: Figure 1In an application environment, the user terminal communicates with the server terminal through a network. The server terminal can collect the reserve resource business data within a preset time range through the user terminal, perform statistical analysis and determine the target analysis model, optimize and update the parameters of the target analysis model based on historical business data, process the reserve resource business data to generate a preliminary analysis result, and apply the adaptive adjustment mode to adjust the preliminary analysis result to generate analysis results under multiple business scenarios. The present invention improves the accuracy and adaptability of the model by determining the model based on data statistical analysis and optimizing and updating the model parameters. Flexible adjustment of multiple business scenarios is achieved through the adaptive adjustment mode, which effectively responds to the liquidity risks of reserve resources under different market conditions and realizes accurate analysis and dynamic management of reserve resource data. Among them, the user terminal can be but is not limited to various personal computers, laptops, smart phones, tablet computers and portable wearable devices. The server terminal can be implemented with an independent server or a server cluster consisting of multiple servers. The present invention is described in detail below through specific embodiments.

[0031] See also Figure 2 , Figure 2 This is a flow chart of an embodiment of a method for analyzing reserve resource business data provided by the present invention. It should be noted that although a logical sequence is shown in the flow chart, in some cases, the steps shown or described may be performed in a different order than that shown here.

[0032] like Figure 2 As shown, the reserve resource business data analysis method proposed by the present invention includes the following steps:

[0033] S10, collecting reserve resource business data within a preset time range;

[0034] In this embodiment, in the process of collecting reserve resource business data, it is first necessary to determine the data source, including the bank's internal database, real-time data stream or third-party data supplier. These data sources may include the core system of the provident fund center, the bank's fund management system, external financial data interface, etc. According to business needs and compliance requirements, connections are established with different data sources through API interfaces, data query languages ​​(such as SQL) or data synchronization tools (such as ETL tools). Data is obtained through periodic batch tasks or real-time stream processing mode.

[0035] Set the time range of data according to the analysis requirements. The time range setting can be flexibly adjusted according to the integrity of historical data and the accuracy requirements of analysis. For example, you can set the data collection range for the last 1 year, the last 5 years, or longer. Set the time window of data through database query conditions or data interface parameters to collect data for a specific time period. The data collection time range setting can also be automatically adjusted according to the analysis cycle of the business system, such as updating data once a quarter, every six months, or every year.

[0036] The collected data may have redundancy, missing values, outliers, etc., so it needs to be cleaned and preprocessed. This includes data deduplication, missing value filling, outlier processing, field mapping, etc. Use data cleaning tools or custom scripts to perform data cleaning and preprocessing operations through statistical methods, machine learning models, etc.

[0037] The collected data needs to be stored according to the preset structure to facilitate the subsequent call and processing of the analysis model. The storage structure should meet the requirements of fast query and efficient reading of data. Relational databases (such as MySQL), data warehouses (such as Hadoop, Hive) or distributed storage systems (such as HDFS) can be used to store collected data. The structural design of data storage must meet the reading requirements of model analysis.

[0038] Example: In the financial field, the collection of reserve resource business data is the basis for banks and other financial institutions to conduct risk assessment and liquidity management. Taking provident fund management as an example, banks regularly collect data on deposits, withdrawals, loan issuance, etc. in the past year from the provident fund center every month, and clean and store these data in the bank's internal data warehouse through batch processing tasks. For periods of large market fluctuations (such as fluctuations in the real estate market), banks can use real-time data streams to monitor the flow of provident fund funds in order to quickly respond to emergencies and make dynamic adjustments. The flexibility and reliability of this data collection method provide accurate data support for banks to formulate reasonable liquidity management strategies.

[0039] By setting a preset time range, the required reserve resource business data can be collected flexibly, ensuring the comprehensiveness of the data while taking into account the timeliness of the data. The data cleaning and preprocessing stages improve the data quality and make the model analysis results more accurate.

[0040] S20, performing statistical analysis on the reserve resource business data, and determining a target analysis model based on the statistical analysis results;

[0041] In this embodiment, a preliminary statistical analysis is performed on the collected reserve resource business data, including the calculation of basic descriptive statistical indicators, such as mean, median, variance, standard deviation, maximum value, minimum value, etc. At the same time, it is also necessary to select appropriate statistical methods according to the type and characteristics of the data, such as time series analysis, regression analysis, etc. Various descriptive statistical indicators are calculated through data analysis tools (such as Python's Pandas, R language, etc.). Basic statistical features are extracted from the reserve resource data through scripts or analysis programs.

[0042] Analyze the distribution characteristics of reserve resource business data, including the time series changes of data, the correlation of data, the volatility of data, etc. The purpose is to identify the trend, cycle and abnormal fluctuation of data, and provide a basis for subsequent model selection. Use time series analysis methods (such as ACF / PACF analysis, unit root test, etc.) to analyze the time characteristics of data; use correlation analysis (such as Pearson correlation coefficient) to determine the correlation between different data variables; and use volatility analysis (such as coefficient of variation, standard deviation, etc.) to determine the fluctuation range of data.

[0043] According to the results of statistical analysis, select the appropriate target analysis model. For example, when the data has obvious time series characteristics, ARIMA or exponential smoothing model may be selected; when there is significant linear correlation between data, regression model may be selected. Use model selection algorithm to automatically or manually screen suitable analysis models based on business needs. The optimal model can be selected from candidate models through automated model evaluation processes (such as AIC / BIC indicator evaluation, cross-validation, etc.).

[0044] Example: In the financial field, statistical analysis of provident fund deposits, withdrawals, and loan issuance data can identify the changing trends and fluctuation patterns of these data. For example, banks can use time series analysis methods to conduct stationarity tests and autocorrelation analysis on the monthly provident fund deposits, and find that the data has a significant cyclical fluctuation trend. On this basis, the bank selects the ARIMA model as the target analysis model to analyze and estimate the deposit trend in the next few months. Through this scientific statistical analysis and model selection process, banks can more accurately grasp the liquidity changes of provident funds and provide data support for formulating reasonable fund management strategies.

[0045] Through detailed statistical analysis of reserve resource business data, the time series characteristics, correlation characteristics and volatility characteristics of the data can be effectively identified, thus providing a scientific basis for the screening of target analysis models. Through the model selection algorithm, the most suitable analysis model can be determined among the candidate models, thereby improving the accuracy and efficiency of the analysis and providing more reliable model support for subsequent liquidity management.

[0046] S30, optimizing and updating the parameters of the target analysis model based on the historical business data of the reserve resources;

[0047] In this embodiment, long-term accumulated data is obtained from the historical business data set of the reserve resources, including but not limited to the monthly reserve resource deposit amount, withdrawal amount, loan issuance and recovery, etc. The purpose of obtaining historical data is to provide a more comprehensive perspective for the parameter optimization of the target analysis model. The required data set is extracted from the historical business data table through database query or batch script, and data cleaning, filtering and preprocessing operations are performed to ensure the integrity and accuracy of the data.

[0048] Based on historical business data, the parameters of the target analysis model are identified and defined. Different types of analysis models have different parameters, such as the order, number of differences, and smoothing coefficients of the ARIMA model; the regression coefficients of the regression model; the smoothing parameters of the exponential smoothing model, etc. Through the model training algorithm, the adjustable parameters in the analysis model (such as the order of the time series model and the smoothing coefficient of the smoothing model) are automatically identified to determine the parameter list that needs to be optimized.

[0049] Use a specific optimization algorithm to optimize the parameters of the target analysis model. Common optimization methods include grid search, random search, gradient descent, genetic algorithm, etc. The goal of optimization is to minimize the model error or residual (such as mean square error, absolute error, etc.) based on historical data. For example, for the ARIMA model, the appropriate order can be selected through the AIC / BIC evaluation index; for the regression model, the regression equation is fitted by the least squares method and the regression coefficient is adjusted; for the exponential smoothing model, the smoothing coefficient is adjusted by minimizing the residual.

[0050] After completing parameter optimization, the optimized model needs to be verified to ensure that the optimized model parameters can improve the accuracy of the analysis. Common verification methods include cross-validation and residual analysis. By backtesting or cross-validating historical data, the performance of the optimized model is evaluated. If the verification results meet expectations, the optimized model parameters are updated and saved.

[0051] Example: In the financial sector, the amount of provident fund deposits and withdrawals is greatly affected by the economic environment and policies. Banks can use the historical business data of the past few years to optimize and update the parameters of the time series model. For example, by obtaining historical provident fund deposit data, banks can use the AIC / BIC criteria and cross-validation methods to determine the optimal order, number of differences, and smoothing coefficient of the ARIMA model. Based on the optimized model, banks can more accurately predict the deposit trend in the next few months, thereby better formulating liquidity management strategies for provident fund funds.

[0052] By using historical business data to optimize and update the parameters of the target analysis model, the accuracy and robustness of the model can be improved and the error of the analysis results can be reduced. The optimized analysis model can better capture the trends and characteristics of the data, thereby providing more reliable support for the liquidity management of reserve resources.

[0053] S40, using the target analysis model to process the reserve resource business data within the preset time range to generate a preliminary analysis result;

[0054] In this embodiment, before the reserve resource business data within the preset time range is input into the target analysis model, the data needs to be standardized and cleaned. The purpose of standardization is to ensure that the scales of different data are consistent to avoid model misjudgment. Use standardization methods (such as Z-score standardization or Min-Max normalization) to process the data so that different types of data features meet the input requirements of the model. At the same time, check the missing values ​​and outliers in the data and fill or remove them.

[0055] For long-term or large-scale reserve resource business data, it is usually necessary to process the data in segments. Segmentation can be divided according to the time dimension of the data or business needs to make the model process more efficiently. The data within the preset time range is segmented by time period (such as by month or quarter), and input into the target analysis model for calculation in batches. Data segmentation can be implemented through data processing libraries such as Python's Pandas.

[0056] Input the standardized and segmented data into the selected and optimized target analysis model for data calculation and analysis. The specific calculation process of the target analysis model depends on the selected model, such as time series prediction of the ARIMA model, correlation analysis of the regression model, trend estimation of the exponential smoothing model, etc. Use data analysis and modeling tools (such as Python's Statsmodels, Scikit-learn, R language, etc.) to run the selected target analysis model, and use the data features as the input variables of the model, and perform calculations by calling the model's prediction or fitting method.

[0057] After the model calculation is completed, the preliminary analysis results containing the trend, volatility and correlation characteristics of the reserve resource data are generated. These analysis results can include information such as estimated data for a period of time in the future, data change trends, fluctuation ranges, etc. The results output by the model are sorted and converted, such as calculating future estimated values, calculating correlation coefficients and fluctuation ranges, etc. Trend graphs, scatter plots, fluctuation graphs, etc. are generated through visualization tools (such as Matplotlib, Seaborn, etc.).

[0058] Example: In the financial sector, data such as provident fund deposits, withdrawals, and loan issuance have significant time series characteristics. After collecting the latest provident fund deposit data every month, banks can standardize these data and segment them by quarter. These data are input into the optimized ARIMA model for time series analysis to generate deposit trend estimates and fluctuation ranges for the next few quarters. Based on these preliminary analysis results, banks can adjust the provident fund's capital reserve strategy to ensure the stability of the provident fund's liquidity, thereby improving the efficiency of fund management.

[0059] By inputting data within a preset time range into the optimized target analysis model, more accurate preliminary analysis results can be generated. The standardization and segmentation processing steps ensure the quality and consistency of data input, thereby improving the calculation efficiency and accuracy of the model. The final preliminary analysis results can provide predictions of future trends and analysis of data fluctuations, providing data support for subsequent business decisions.

[0060] S50, adjusting the preliminary analysis result through an adaptive adjustment mode to generate analysis results in multiple business scenarios.

[0061] In this embodiment, before adjusting the preliminary analysis results, it is necessary to define multiple business scenarios according to different business needs, and each scenario may contain different adjustment targets and preset conditions. For example, different business scenarios may require maintaining a certain liquidity or responding to specific market fluctuations in different time periods. Define different business scenarios through business rules or user demand analysis. Parameterize the adjustment target of each business scenario, such as setting the liquidity target of a certain business scenario to 50% or setting the risk tolerance under different business scenarios.

[0062] Select the appropriate adaptive adjustment mode based on the adjustment target of each business scenario and the data features in the preliminary analysis results. Adaptive adjustment modes can include proportional adjustment mode, fixed value adjustment mode, smooth adjustment mode, and arithmetic adjustment mode. Select the adaptive adjustment mode through predefined business logic and model algorithms (such as conditional judgment, rule selection, or intelligent algorithm). For example, when the fluctuation of the preliminary analysis results exceeds the preset range, the smooth adjustment mode can be selected; when the business needs require a fixed adjustment value, the fixed value adjustment mode can be selected.

[0063] Set the corresponding adjustment parameters for the selected adaptive adjustment mode. For example, for the proportional adjustment mode, the proportional parameter needs to be set; for the fixed value adjustment mode, the fixed adjustment value needs to be set; for the smoothing adjustment mode, the smoothing coefficient needs to be set, etc. Input the corresponding adjustment parameters through the user interface or program interface. The optimal value of the adjustment parameter can also be automatically calculated through an optimization algorithm (such as grid search, genetic algorithm).

[0064] After setting the adjustment parameters, the selected adaptive adjustment mode is applied to adjust the preliminary analysis results to generate analysis results in multiple business scenarios. The adjustment process should maintain the continuity and consistency of the data. Through the algorithm logic of the adjustment mode, the key data in the preliminary analysis results are adjusted accordingly in numerical value or trend. The mathematical operation library in Python or R is applied to realize the adjustment of different modes.

[0065] Generate analysis results that meet the needs of each business scenario based on different adjustment modes and parameters. These analysis results should include key indicators, trend changes, volatility and other information for each business scenario. Sort and summarize the adjusted data for each business scenario, and display the results of each business scenario through the data repository or analysis report system. You can generate different charts or analysis reports to intuitively view the changes in different business scenarios.

[0066] Example description: In the financial field, for example, when a bank manages the liquidity of a provident fund, it can define multiple business scenarios according to market changes, such as market stability period, market volatility period and market abnormal volatility period. During the market stability period, the bank may want to keep the liquidity of the provident fund at a high level, so it chooses the proportional adjustment mode to maintain a certain proportion of capital flow. During the market volatility period, the bank may want to reduce the impact of volatility on liquidity, so it uses the smoothing adjustment mode to reduce the impact of volatility. During the abnormal market volatility period, the bank may need to adopt a more conservative strategy, so it uses the fixed value adjustment mode to ensure that liquidity remains within the set safety range. Through these adaptive adjustment modes, banks can flexibly respond to different market environments and optimize fund management strategies.

[0067] Through the adaptive adjustment mode, the preliminary analysis results can be flexibly adjusted to meet the needs of different business scenarios. Each adaptive adjustment mode makes fine parameter adjustments for specific business scenarios, making the analysis results more targeted and accurate. This flexible adjustment method can help managers better respond to market fluctuations and economic changes and formulate appropriate business strategies and capital allocation plans.

[0068] The present invention relates to the fields of data processing technology and financial technology, and discloses a method for analyzing reserve resource business data, which collects reserve resource business data within a preset time range, performs statistical analysis and determines a target analysis model, optimizes and updates the parameters of the target analysis model based on historical business data, processes the reserve resource business data to generate preliminary analysis results, and applies an adaptive adjustment mode to adjust the preliminary analysis results to generate analysis results under multiple business scenarios. The present invention improves the accuracy and adaptability of the model by determining a model based on data statistical analysis and optimizing and updating the model parameters. Flexible adjustment of multiple business scenarios is achieved through the adaptive adjustment mode, which effectively responds to the reserve resource liquidity risks under different market conditions, and realizes accurate analysis and dynamic management of reserve resource data.

[0069] In one embodiment, the above S30 includes:

[0070] S301, if the target analysis model is a time series model, optimizing and adjusting the order, number of differences and smoothing coefficient of the time series model based on the historical business data of the reserve resources;

[0071] S302, if the target analysis model is a regression model, fitting a regression equation based on historical business data of the reserve resources to adjust the regression coefficient in the regression model;

[0072] S303: If the target analysis model is an exponential smoothing model, then based on the historical business data of the reserve resources, the smoothing coefficient of the exponential smoothing model is updated.

[0073] In this embodiment, common time series models include ARIMA model (autoregressive integrated moving average model) and SARIMA model (seasonal autoregressive integrated moving average model). In the time series model, the main parameters that need to be optimized include the autoregressive order, the number of differences and the smoothing coefficient of the model. By performing time series analysis on the historical business data of the reserve resources, the model selection method (such as AIC, BIC criteria) is used to determine the autoregressive order, the number of differences and the smoothing coefficient. For the ARIMA model, the following method can be used to adjust the parameters:

[0074] Autoregressive order (AR): Use ACF and PACF plots to determine the autoregressive order of the data.

[0075] Difference number (I): Based on the ADF (Augmented Dickey-Fuller) unit root test results, determine the number of differences in the data to ensure stationarity.

[0076] Smoothing coefficient (MA): Use residual analysis to determine the optimal value of the smoothing coefficient.

[0077] Common regression models include linear regression, multiple regression, ridge regression, etc. In regression analysis, the coefficient of the regression equation (i.e., regression coefficient) is the key parameter that needs to be optimized and adjusted. Based on historical business data, the least squares method (OLS) is used to fit the regression equation and adjust the regression coefficient to minimize the sum of squared errors. The following steps can be used:

[0078] Build a regression model: According to the characteristics of business data, select a single variable or multivariate regression model and build a regression equation.

[0079] Calculate regression coefficients: Calculate regression coefficients by the method of least squares and verify the significance of the coefficients (such as by t-test or p-value).

[0080] Optimization adjustment: According to the distribution of residuals and the goodness of fit of the model (such as R 2 ) to judge the accuracy of the regression model and make further adjustments to the regression coefficients.

[0081] Common exponential smoothing models include simple exponential smoothing (SES), double exponential smoothing, and triple exponential smoothing (Holt-Winters). In the exponential smoothing model, the smoothing coefficients (α, β, γ) determine the degree of data smoothing. Based on historical business data, the smoothing coefficients are adjusted by minimizing the forecast error (such as mean square error MSE, mean absolute error MAE). The specific methods are as follows:

[0082] Set initial smoothing coefficient: Set the initial smoothing coefficient α for single exponential smoothing, or set the smoothing coefficient α and trend coefficient β for double exponential smoothing.

[0083] Perform error analysis: perform multiple model calculations using different smoothing coefficients, evaluate the error of each calculation, and select the optimal smoothing coefficient.

[0084] Apply optimization algorithms: Automatically calculate the optimal smoothing coefficient through grid search, genetic algorithm, etc.

[0085] This embodiment can significantly improve the accuracy and applicability of various types of analysis models (time series model, regression model, exponential smoothing model) by optimizing parameters for different types of analysis models. The optimization of the order, number of differences and smoothing coefficient of the time series model helps to better capture the time series characteristics of the reserve resource data, the optimization of the regression coefficient of the regression model can more accurately reflect the linear relationship between variables, and the adjustment of the smoothing coefficient of the exponential smoothing model can better smooth data fluctuations. Overall, by optimizing the model parameters, the prediction accuracy and applicability of the analysis model can be improved, ensuring the reliability of the analysis results.

[0086] In one embodiment, the above S20 includes:

[0087] S201, performing a preprocessing operation on the reserve resource service data;

[0088] S202, performing statistical analysis on the reserve resource business data after the preprocessing operation, obtaining basic statistical indicators, and analyzing the distribution characteristics of the reserve resource business data;

[0089] S203, analyzing and obtaining the time series characteristics, correlation characteristics and volatility characteristics of the reserve resource business data based on the basic statistical indicators and distribution characteristics;

[0090] S204, determining a target analysis model based on the time series characteristics, correlation characteristics and volatility characteristics.

[0091] In this embodiment, data preprocessing includes steps such as cleaning the original reserve resource business data, format unification, missing value filling, data deduplication, etc. to ensure the integrity and consistency of the data. The purpose of preprocessing is to improve data quality and reduce the impact of abnormal data on statistical analysis. Use data cleaning techniques and tools (such as Python's Pandas library, R language data cleaning package, etc.) to process the original data. Data preprocessing can be performed by setting missing value filling rules (such as mean filling, interpolation, etc.) and outlier detection algorithms (such as Z-score method).

[0092] Statistical analysis includes the calculation of descriptive statistical indicators of the data (such as mean, median, standard deviation, variance, maximum value, minimum value, etc.), as well as the analysis of data distribution (such as normality test, skewness and kurtosis analysis, etc.). The purpose of the analysis is to understand the basic characteristics and distribution of the data. Basic statistical indicators are calculated through data analysis tools (such as Python's statistical package SciPy, R language's descriptive statistical functions, etc.). At the same time, graphical tools (such as Matplotlib, Seaborn, etc.) can be used to generate data distribution graphs (such as histograms, QQ graphs) to further analyze the distribution characteristics of the data.

[0093] Perform time series analysis on reserve resource business data to identify trends, cyclical and seasonal changes in the data. Time series decomposition methods can be used to extract these features. Use correlation analysis (such as Pearson correlation coefficient, Spearman rank correlation coefficient, etc.) to determine the linear or nonlinear relationship between different data variables. Assess the volatility of the data and use statistical indicators such as standard deviation and coefficient of variation to describe the fluctuation range of the data.

[0094] Time series feature analysis can use methods such as ACF / PACF diagrams and autocorrelation analysis. Correlation feature analysis can be achieved through correlation matrix or correlation scatter plot. Volatility feature analysis can calculate volatility indicators such as standard deviation and coefficient of variation through statistical methods.

[0095] Select the most appropriate target analysis model based on the results of the above feature analysis. For example, when the time series features are more obvious, you can choose the ARIMA model; when the correlation features are significant, you can choose the regression model; when the data fluctuates frequently, you can choose the exponential smoothing model, etc. Evaluate the suitability of candidate models through model selection algorithms (such as AIC / BIC criteria, cross-validation, etc.) to determine the best target analysis model. Automated model selection tools (such as Python's statsmodels, Scikit-learn, etc.) can be used for model evaluation and selection.

[0096] This embodiment can effectively extract the time series characteristics, correlation characteristics and volatility characteristics of the data by preprocessing and performing detailed statistical analysis on the reserve resource business data, thereby providing a scientific basis for the selection of the target analysis model. This not only improves the accuracy of model selection, but also enhances the flexibility of the model, so that the model can better adapt to the characteristics of the reserve resource data, thereby improving the accuracy and efficiency of the analysis.

[0097] In one embodiment, after the above S30, the method further includes:

[0098] S304, obtaining the latest reserve resource business data at a preset time interval;

[0099] S305, retraining and updating parameters of the target analysis model based on the latest reserve resource business data;

[0100] S306, verifying the analysis effect of the retrained target analysis model through the performance evaluation module, and adjusting the parameters of the target analysis model according to the verification result.

[0101] In this embodiment, according to the actual needs of the business and the accuracy requirements of the analysis, a reasonable time interval is set to regularly obtain the latest reserve resource business data. The time interval can be daily, weekly, monthly, etc., depending on the characteristics of the business. The data acquisition operation is regularly triggered through a batch task or scheduling system (such as CronJob, Apache Airflow), and the latest reserve resource business data is extracted from the business data source (such as a database, data lake, API, etc.) and stored in the data warehouse or analysis platform.

[0102] Retrain the target analysis model using the latest data to update the model's weights and parameters. The purpose of retraining is to adapt the model to the latest data changes and maintain the model's prediction accuracy. During the retraining process, optimize the model's parameters (such as the order of the ARIMA model, the coefficients of the regression model, the smoothing coefficients of the exponential smoothing model, etc.) to ensure that the retrained model achieves the best results on the latest data. Retrain the model on the latest reserve resource business data by calling machine learning libraries (such as Python's Scikit-learn, TensorFlow, etc.) or statistical modeling tools. The parameters of the model can be optimized and updated by methods such as batch gradient descent, least squares, or genetic algorithms.

[0103] Use the performance evaluation module to evaluate the performance of the system through evaluation indicators such as mean square error (MSE), mean absolute error (MAE), R 2 The purpose is to ensure that the retrained model can maintain high accuracy and adapt to the latest data. According to the performance evaluation results, if the performance of the retrained model does not meet expectations, the model parameters need to be further adjusted to optimize the performance of the model.

[0104] After the model is retrained, the validation set is used to evaluate the model's performance, and the various performance indicators of the model are calculated by writing a validation script. If the evaluation results show that the error is too large, the parameters can be further optimized by optimizing the algorithm, adjusting the model structure, etc.

[0105] This embodiment obtains the latest business data at preset time intervals and regularly retrains and updates the parameters of the analysis model, so that the model can always maintain sensitivity and adaptability to the latest data. The performance evaluation module verifies the effect of the model, and can flexibly adjust parameters to adapt to different business scenarios while ensuring the accuracy of the model.

[0106] In one embodiment, the above S40 includes:

[0107] S401, performing standardization processing on the reserve resource business data;

[0108] S402, segmenting the standardized reserve resource business data;

[0109] S403, inputting the segmented reserve resource business data into the target analysis model;

[0110] S404, analyzing the trend characteristics, correlation characteristics and volatility characteristics of the reserve resource business data through the target analysis model, and generating a preliminary analysis result including reserve resource trend, estimated value and volatility information.

[0111] In this embodiment, standardization refers to normalizing or removing bias from the reserve resource business data to ensure that different feature data are in the same dimension and reduce the impact of data in different units or magnitudes on the analysis model. Common standardization methods include Z-score standardization, Min-Max normalization, etc. The data is normalized using a standardization algorithm. For example, for Z-score standardization, the formula (x-μ) / σ can be used for conversion, where x is the data value, μ is the mean, and σ is the standard deviation. This can be implemented using Python's Scikit-learn library or the scale function in R language.

[0112] Segmentation refers to dividing the reserve resource business data by time period or business cycle, so that the subsequent input analysis model can better reflect the characteristics of different time periods or business changes. The basis for segmentation can be natural time periods (such as months, quarters) or business change cycles (such as different market conditions). Use data processing scripts (such as Python's Pandas library) to segment the standardized data by time period or by business changes. Data segmentation by month, quarter, or specific business event can be achieved by writing a data segmentation function.

[0113] The segmented data is input into the determined and optimized target analysis model in batches according to the input format of the model. The target analysis model can be a time series model, a regression model, or an exponential smoothing model. By programming and calling the API interface or custom function of the target analysis model (such as ARIMA, regression model, exponential smoothing model), the segmented data is input into the model in sequence for calculation.

[0114] The target analysis model is used to analyze three characteristics of the data, namely trend characteristics (long-term trend), correlation characteristics (correlation with other variables) and volatility characteristics (data fluctuation range and instability). The goal of the analysis is to generate a preliminary analysis result containing information such as future data forecasts and fluctuation ranges. Different target analysis models may have different analysis methods for different characteristics. For example, the ARIMA model can be used to analyze long-term trends and smooth fluctuations, the regression model is used to analyze the correlation between variables, and the exponential smoothing model is used to predict the fluctuation range of data. The analysis of features and the generation of results can be achieved through data analysis tools (such as Python's Statsmodels and Scikit-learn).

[0115] This embodiment, through the gradual application of standardized processing, segmented processing and target analysis models, can make data more consistently processed and analyzed in different models, reducing errors and fluctuations in model calculations. By analyzing trend characteristics, correlation characteristics and volatility characteristics, it can more accurately reflect the long-term change trend, potential influencing factors and fluctuation range of reserve resource business data, provide managers with more comprehensive preliminary analysis results, and thus formulate more sophisticated fund management strategies.

[0116] In one embodiment, the above S50 includes:

[0117] S501, setting adjustment targets and preset conditions for multiple business scenarios according to business needs;

[0118] S502, determining a target adaptive adjustment mode based on the adjustment target and preset conditions of each business scenario and in combination with the data features in the preliminary analysis result;

[0119] S503, setting corresponding adjustment parameters for the target adaptive adjustment mode;

[0120] S504: Adjust the preliminary analysis result based on the target adaptive adjustment mode to generate an analysis result for each business scenario.

[0121] In this embodiment, multiple business scenarios are set according to different business needs and goals, and adjustment goals and preset conditions are defined for each scenario. Adjustment goals refer to specific indicators or states that are expected to be achieved in each business scenario (such as liquidity level, fund allocation ratio, etc.), and preset conditions refer to business constraints or external conditions that affect these goals (such as market volatility, economic policy changes, etc.). Through business demand analysis and pattern recognition of historical data, multiple typical business scenarios are defined, such as market stability scenarios, market volatility scenarios, emergency scenarios, etc. The adjustment goals and preset conditions in each scenario can be set by business experts or automatically generated by machine learning algorithms.

[0122] According to the adjustment goals and preset conditions in each business scenario, combined with the data characteristics in the preliminary analysis results (such as trends, correlations, and volatility characteristics), determine the appropriate adaptive adjustment mode. Different adjustment modes can include proportional adjustment, fixed value adjustment, smoothing adjustment, etc. According to business rules or adaptive adjustment strategies, select the target adaptive adjustment mode by judging the relationship between the adjustment goals and data characteristics. For example, when the data fluctuates violently, select the smoothing adjustment mode to reduce the impact of fluctuations; when the market trend changes slightly but requires a small adjustment, select the fixed value adjustment mode.

[0123] According to the selected adaptive adjustment mode, set the corresponding adjustment parameters. Different adjustment modes require different parameters, such as the proportional coefficient of the proportional adjustment mode, the adjustment value of the fixed value adjustment mode, and the smoothing coefficient of the smoothing adjustment mode. The adjustment parameters can be determined by business settings or automatic calculation by algorithms. The parameter adjustment algorithm (such as grid search and random search) can be used to optimize the value of the adjustment parameters so that the adjustment effect is the best in each business scenario.

[0124] Apply the target adaptive adjustment mode to adjust the data in the preliminary analysis results numerically or trend-adjust to generate analysis results that meet the needs of each business scenario. The adjusted analysis results should include information such as key indicators, change trends, and fluctuation characteristics in each business scenario. Adjustment operations are implemented through programming, such as using mathematical operation libraries in Python or R to implement proportional adjustment, smoothing adjustment, etc., and the analysis results in each business scenario are displayed through a data repository or visualization system.

[0125] Example: In the financial field, for example, when a bank manages the liquidity of a provident fund, it can set multiple business scenarios according to market changes, such as market stability scenarios, market volatility scenarios, and policy change scenarios. For the market stability scenario, the bank can set a liquidity target of 90% and select the proportional adjustment mode to adjust all preliminary analysis results to the target level in proportion. In the market volatility scenario, the bank sets a smoothing adjustment mode and sets a smoothing coefficient based on the volatility characteristics to reduce the impact of volatility through smoothing adjustments. Through this multi-scenario adaptive adjustment, banks can better respond to different market environments and optimize liquidity management.

[0126] This embodiment can flexibly respond to various business changes by defining adjustment targets and preset conditions for different business scenarios, combining the data features in the preliminary analysis results, selecting a suitable adaptive adjustment mode, and setting corresponding adjustment parameters. Each adaptive adjustment mode makes fine parameter adjustments for specific business scenarios, so that the adjusted analysis results are more targeted and accurate. This flexible adjustment method helps managers better respond to market fluctuations and external changes, and formulate reasonable fund management and strategy adjustment plans.

[0127] In one embodiment, after the above S50, the method further includes:

[0128] S601, classifying the analysis results according to the types, time periods and correlation characteristics of the reserve resources to obtain data classification results;

[0129] S602, based on the data classification result, generating a corresponding visualization chart to display the change trend, fluctuation and characteristic analysis results of the reserve resources;

[0130] S603: Generate an overall analysis report based on the visualization chart.

[0131] In this embodiment, data classification refers to systematically classifying and arranging the analysis results according to the types of different reserve resources (such as provident funds, deposits, loans, etc.), as well as the time period (such as monthly, quarterly, annual) and the correlation characteristics of the data. The purpose of classification is to make subsequent analysis and presentation clearer. Through programming or using data processing tools (such as Python's Pandas library, SQL database, etc.), the analysis results are grouped and sorted according to classification rules. Custom classification algorithms or rules can be written to classify data.

[0132] Based on the data classification results, a series of visualization charts are generated to intuitively display the changing trends, fluctuations, and correlation characteristics of reserve resources. The types of visualization charts can include trend line charts, fluctuation bar charts, correlation scatter plots, heat maps, etc. Various charts are generated using data visualization tools (such as Python's Matplotlib, Seaborn, Plotly, etc.). By calling the corresponding visualization functions and setting parameters, the changes and data characteristics of different reserve resource types in different time periods can be displayed.

[0133] The classified data and visualization charts are organized into an overall analysis report, which should include the changing trend, fluctuation, characteristic analysis results, etc. of the reserve resources. The purpose of the analysis report is to provide comprehensive and accurate data support for decision-makers. Visualization charts and data analysis results are integrated through report generation tools (such as Python's ReportLab, LaTeX, etc.) to generate report files in PDF format or other standard formats. The report should include descriptions of various charts, data summaries, and analysis conclusions.

[0134] Example: In the financial field, banks can classify different types of provident fund data by monthly, quarterly, and annual basis for the management of provident funds. Visualization tools are used to generate a line chart of the quarterly trend of provident funds, a volatility bar chart, and a heat map of the correlation between provident funds and economic factors. Finally, the bank integrates the visualization charts and analysis results to generate an overall analysis report, which shows the changing trends and volatility of different reserve resources in various business scenarios. In this way, bank managers can clearly grasp the liquidity status of various reserve resources and provide decision support for the formulation of scientific fund management strategies.

[0135] This embodiment can make data management and analysis more systematic and organized by classifying the analysis results, which helps to present clearer business logic in displays and reports. The generated visualization charts can intuitively display the changing trends, fluctuations and characteristic analysis results of reserve resources, providing managers with intuitive decision-making basis. The overall analysis report finally generated can comprehensively and accurately summarize and summarize various data analysis results, providing a scientific basis for business optimization and resource allocation.

[0136] In one embodiment, a reserve resource business data analysis device is provided, and the reserve resource business data analysis device corresponds one-to-one to the reserve resource business data analysis method in the above embodiment. Figure 3 , Figure 3 This is a functional module diagram of a preferred embodiment of the resource reserve business data analysis device of the present invention. Data collection module 10, data analysis and model selection module 20, model optimization and training module 30, model application and analysis module 40 and result adjustment and scenario management module 50. The functional modules are described in detail as follows:

[0137] The data collection module 10 is used to collect the reserve resource business data within a preset time range;

[0138] A data analysis and model selection module 20, configured to perform statistical analysis on the reserve resource business data and determine a target analysis model based on the statistical analysis results;

[0139] A model optimization and training module 30, for optimizing and updating the parameters of the target analysis model based on the historical business data of the reserve resources;

[0140] A model application and analysis module 40, configured to use the target analysis model to process the reserve resource business data within the preset time range to generate a preliminary analysis result;

[0141] The result adjustment and scenario management module 50 is used to adjust the preliminary analysis results through an adaptive adjustment mode to generate analysis results in multiple business scenarios.

[0142] In one embodiment, the model optimization and training module 30 is specifically used to:

[0143] If the target analysis model is a time series model, the order, number of differences and smoothing coefficient of the time series model are optimized and adjusted based on the historical business data of the reserve resources;

[0144] If the target analysis model is a regression model, a regression equation is fitted based on the historical business data of the reserve resources to adjust the regression coefficient in the regression model;

[0145] If the target analysis model is an exponential smoothing model, the smoothing coefficient of the exponential smoothing model is updated based on the historical business data of the reserve resources.

[0146] In one embodiment, the data analysis and model selection module 20 is specifically used to:

[0147] Performing preprocessing operations on the reserve resource business data;

[0148] Performing statistical analysis on the reserve resource business data after the preprocessing operation, obtaining basic statistical indicators, and analyzing the distribution characteristics of the reserve resource business data;

[0149] Based on the basic statistical indicators and distribution characteristics, analyzing and obtaining the time series characteristics, correlation characteristics and volatility characteristics of the reserve resource business data;

[0150] A target analysis model is determined based on the time series characteristics, correlation characteristics and volatility characteristics.

[0151] In one embodiment, the model optimization and training module 30 is specifically used to:

[0152] Obtain the latest reserve resource business data at preset time intervals;

[0153] Based on the latest reserve resource business data, retraining and updating parameters of the target analysis model;

[0154] The analysis effect of the retrained target analysis model is verified through the performance evaluation module, and the parameters of the target analysis model are adjusted according to the verification results.

[0155] In one embodiment, the model application and analysis module 40 is specifically used to:

[0156] Performing standardization processing on the reserve resource business data;

[0157] Segment processing of the standardized reserve resource business data;

[0158] Inputting the segmented reserve resource business data into the target analysis model;

[0159] The target analysis model is used to analyze the trend characteristics, correlation characteristics and volatility characteristics of the reserve resource business data, and generate preliminary analysis results including reserve resource trend, estimated value and volatility information.

[0160] In one embodiment, the result adjustment and scene management module 50 is specifically used to:

[0161] Set adjustment targets and preset conditions for multiple business scenarios based on business needs;

[0162] Determine a target adaptive adjustment mode based on the adjustment target and preset conditions of each business scenario and in combination with the data features in the preliminary analysis results;

[0163] Setting corresponding adjustment parameters for the target adaptive adjustment mode;

[0164] The preliminary analysis result is adjusted based on the target adaptive adjustment mode to generate an analysis result for each business scenario.

[0165] In one embodiment, the result adjustment and scene management module 50 is specifically used to:

[0166] Classifying the analysis results according to the types, time periods and correlation characteristics of the reserve resources to obtain data classification results;

[0167] Based on the data classification results, generate corresponding visualization charts to show the change trend, fluctuation and characteristic analysis results of reserve resources;

[0168] Based on the visualization chart, an overall analysis report is generated.

[0169] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 4 As shown. The computer device includes a processor, a memory, a network interface and a database connected via a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile and / or volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external user terminal through a network connection. When the computer program is executed by the processor, it realizes the functions or steps on the service side of a reserve resource business data analysis method.

[0170] In one embodiment, a computer device is provided. The computer device may be a user terminal, and its internal structure diagram may be as follows: Figure 5As shown. The computer device includes a processor, a memory, a network interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external server via a network connection. When the computer program is executed by the processor, it implements the functions or steps of a user-side method for analyzing reserve resource business data

[0171] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the following steps are implemented:

[0172] Collect reserve resource business data within a preset time range;

[0173] Performing statistical analysis on the reserve resource business data, and determining a target analysis model based on the statistical analysis results;

[0174] Based on the historical business data of the reserve resources, the parameters of the target analysis model are optimized and updated;

[0175] Using the target analysis model to process the reserve resource business data within the preset time range to generate a preliminary analysis result;

[0176] The preliminary analysis results are adjusted through an adaptive adjustment mode to generate analysis results in multiple business scenarios.

[0177] In one embodiment, a computer readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

[0178] Collect reserve resource business data within a preset time range;

[0179] Performing statistical analysis on the reserve resource business data, and determining a target analysis model based on the statistical analysis results;

[0180] Based on the historical business data of the reserve resources, the parameters of the target analysis model are optimized and updated;

[0181] Using the target analysis model to process the reserve resource business data within the preset time range to generate a preliminary analysis result;

[0182] The preliminary analysis results are adjusted through an adaptive adjustment mode to generate analysis results in multiple business scenarios.

[0183] It should be noted that the above functions or steps that can be implemented by the computer-readable storage medium or computer device can refer to the relevant descriptions on the server side and the user side in the aforementioned method embodiment. To avoid repetition, they will not be described one by one here.

[0184] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0185] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0186] It should be noted that if software tools or components other than those of the Company appear in the embodiments of the present application, they are only used for illustration and do not represent actual use. The above-described embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the above-mentioned embodiments, a person of ordinary skill in the art should understand that the technical solutions described in the above-mentioned embodiments can still be modified, or some of the technical features thereof can be replaced by equivalents; and these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A method for analyzing reserve resource business data, characterized in that: The following steps are involved: Collect reserve resource business data within a preset time range; Performing statistical analysis on the reserve resource business data, and determining a target analysis model based on the statistical analysis results; Based on the historical business data of the reserve resources, the parameters of the target analysis model are optimized and updated; Using the target analysis model to process the reserve resource business data within the preset time range to generate a preliminary analysis result; The preliminary analysis results are adjusted through an adaptive adjustment mode to generate analysis results in multiple business scenarios.

2. The method for analyzing reserve resource business data according to claim 1, characterized in that: Based on the historical business data of the reserve resources, the parameters of the target analysis model are optimized and updated, including: If the target analysis model is a time series model, the order, number of differences and smoothing coefficient of the time series model are optimized and adjusted based on the historical business data of the reserve resources; If the target analysis model is a regression model, a regression equation is fitted based on the historical business data of the reserve resources to adjust the regression coefficient in the regression model; If the target analysis model is an exponential smoothing model, the smoothing coefficient of the exponential smoothing model is updated based on the historical business data of the reserve resources.

3. The method for analyzing reserve resource business data according to claim 1, characterized in that: Performing statistical analysis on the reserve resource business data and determining a target analysis model based on the statistical analysis results includes: Performing preprocessing operations on the reserve resource business data; Performing statistical analysis on the reserve resource business data after the preprocessing operation, obtaining basic statistical indicators, and analyzing the distribution characteristics of the reserve resource business data; Based on the basic statistical indicators and distribution characteristics, analyzing and obtaining the time series characteristics, correlation characteristics and volatility characteristics of the reserve resource business data; A target analysis model is determined based on the time series characteristics, correlation characteristics and volatility characteristics.

4. The method for analyzing reserve resource business data as claimed in claim 1, characterized in that: After optimizing and updating the parameters of the target analysis model based on the historical business data of the reserve resources, the method further includes: Obtain the latest reserve resource business data at preset time intervals; Based on the latest reserve resource business data, retraining and updating parameters of the target analysis model; The analysis effect of the retrained target analysis model is verified through the performance evaluation module, and the parameters of the target analysis model are adjusted according to the verification results.

5. The method for analyzing reserve resource business data according to claim 1, characterized in that: Using the target analysis model to process the reserve resource business data within the preset time range to generate preliminary analysis results, including: Performing standardization processing on the reserve resource business data; Segment processing of the standardized reserve resource business data; Inputting the segmented reserve resource business data into the target analysis model; The target analysis model is used to analyze the trend characteristics, correlation characteristics and volatility characteristics of the reserve resource business data, and generate preliminary analysis results including reserve resource trend, estimated value and volatility information.

6. The method for analyzing reserve resource business data according to claim 1, characterized in that: The preliminary analysis results are adjusted through the adaptive adjustment mode to generate analysis results in multiple business scenarios, including: Set adjustment targets and preset conditions for multiple business scenarios based on business needs; Determine a target adaptive adjustment mode based on the adjustment target and preset conditions of each business scenario and in combination with the data features in the preliminary analysis results; Setting corresponding adjustment parameters for the target adaptive adjustment mode; The preliminary analysis result is adjusted based on the target adaptive adjustment mode to generate an analysis result for each business scenario.

7. The method for analyzing reserve resource business data according to claim 1, characterized in that: After the preliminary analysis result is adjusted by the adaptive adjustment mode to generate analysis results in multiple business scenarios, the method further includes: Classifying the analysis results according to the types, time periods and correlation characteristics of the reserve resources to obtain data classification results; Based on the data classification results, generate corresponding visualization charts to show the change trend, fluctuation and characteristic analysis results of reserve resources; Based on the visualization chart, an overall analysis report is generated.

8. A reserve resource business data analysis device, characterized in that: The reserve resource business data analysis device comprises: A data collection module is used to collect the reserve resource business data within a preset time range; A data analysis and model selection module, used to perform statistical analysis on the reserve resource business data and determine a target analysis model based on the statistical analysis results; A model optimization and training module, used to optimize and update the parameters of the target analysis model based on the historical business data of the reserve resources; A model application and analysis module, used to use the target analysis model to process the reserve resource business data within the preset time range to generate preliminary analysis results; The result adjustment and scenario management module is used to adjust the preliminary analysis results through an adaptive adjustment mode to generate analysis results in multiple business scenarios.

9. A computer device, characterized in that: The computer device includes a memory, a processor, and a reserve resource business data analysis program stored in the memory and executable on the processor. When the reserve resource business data analysis program is executed by the processor, the steps of the reserve resource business data analysis method as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, characterized in that: The storage medium stores a reserve resource business data analysis program, which, when executed by a processor, implements the steps of the reserve resource business data analysis method according to any one of claims 1 to 7.

Images