Fintech and green finance coupling analysis method and system

Abstract

The present invention provides a method and system for analyzing the coupling between financial technology and green finance, and relates to the technical field of green finance analysis. The present invention provides a dynamic and comprehensive coupling analysis method by quantitatively analyzing policy information and corporate performance. By comprehensively considering the impact of multiple factors such as policy impact, corporate production and sales data, etc. on corporate revenue, the performance indicators of enterprises in environmental protection and sustainable development can be effectively measured. Compared with existing methods, the objectivity and accuracy of policy evaluation are improved, and the actual impact of policy changes can be reflected in a timely manner. At the same time, combined with time series and regression models, the accuracy and flexibility of revenue forecasting are enhanced, providing strong technical support for corporate green development decisions, ensuring that enterprises can respond to changes in the external environment in a timely manner, so as to help enterprises better allocate and optimize resources and improve operational efficiency and economic benefits.

Description

Technical Field

[0001] The present invention relates to the field of green finance analysis technology, and specifically to a method and system for analyzing the coupling between financial technology and green finance. Background Art

[0002] Green finance refers to an economic model that supports environmental protection and sustainable development through financial activities. It includes providing financing support for environmental protection projects, renewable energy, energy conservation and emission reduction, and low-carbon technologies, aiming to promote the green transformation of the economy and environmental protection. The goal of green finance is to promote efficient use of resources, mitigate the impact of climate change on the economy, and achieve sustainable economic growth.

[0003] However, current green finance analysis methods are mostly limited to qualitative assessments and lack systematic quantitative methods. Existing technologies often rely on expert opinions and empirical judgments, and the evaluation results of policy impacts and corporate performance are subjective and highly uncertain. In addition, existing analysis tools do not have the ability to make real-time dynamic adjustments, making it difficult to cope with rapid changes in the policy environment and unable to provide timely and accurate decision-making support for enterprises.

[0004] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not constitute the prior art that is already known to one of ordinary skill in the art. Summary of the invention

[0005] The purpose of the present invention is to provide a method and system for analyzing the coupling between financial technology and green finance to solve the problems raised in the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] The analysis method of coupling between financial technology and green finance includes the following specific steps:

[0008] S1: Collect policy information related to the environment over the years, historical output information and historical revenue data of enterprises;

[0009] S2: Build a text recognition model to process policy information from previous years, quantify and score different clauses in the policy information, and generate corresponding policy impact indexes;

[0010] S3: Generate the green performance index of enterprises on the environment based on the policy impact index and historical output information, and build a correlation model between the green performance index and historical revenue data;

[0011] S4: Build a time series algorithm model to conduct time series analysis on the company's historical output information and generate the forecast output information for the current year;

[0012] S5: Collect the policy information related to the environment in the current year, generate a forecast performance index based on the policy information and forecast output information of the current year, and then forecast the revenue data of the current year based on the forecast performance index.

[0013] Preferably, the policy information includes regulatory documents, subsidy documents, and tax documents; the historical output information includes historical production volume and historical sales volume; and the predicted output information includes predicted production volume and predicted sales volume.

[0014] Preferably, the text recognition model uses a preset natural language processing model, and the logic of quantitatively scoring different clauses in the policy information is:

[0015] Clean the different documents in the policy information, remove the headers, footers, page numbers, and appendices, retain only the document body, and convert the document body into UTF-8 encoding format;

[0016] Split the document body into phrases, remove stop words in the document body according to the preset stop word library, and convert different phrases into stem forms;

[0017] Calculate the frequency of occurrence of each phrase in the document body, and generate a corresponding importance index based on the frequency of occurrence of each phrase;

[0018] Based on the importance index of each phrase in the document body, a policy relevance index and a corresponding relevance weight are generated, and a policy impact index of the policy information is generated according to the policy relevance index and the relevance weight.

[0019] Preferably, the frequency of occurrence of each phrase in the document body is calculated as follows:

[0020] ;

[0021] In the formula Indicates The phrase in The frequency of occurrence in the body of the document, Indicates The phrase in The number of occurrences in the body of the document, Indicates The total number of phrases in the body of the document;

[0022] The importance index of each phrase is calculated as follows:

[0023] ;

[0024] In the formula Indicates The phrase in The importance index in the body of the document, Indicates that it contains The number of documents with the phrase, Indicates the total number of documents;

[0025] The policy relevance index for each document body is calculated as:

[0026] ;

[0027] In the formula Indicates The policy relevance index of the document body;

[0028] The relevance weight of each document body is calculated as:

[0029] ;

[0030] The policy impact index of policy information is calculated as follows:

[0031] ;

[0032] In the formula Represents the policy impact index.

[0033] Preferably, the generation logic of the green performance index is:

[0034] The historical output information is normalized and the calculation methods are as follows:

[0035] ;

[0036] ;

[0037] In the formula , Respectively represent Annual historical production volume and historical sales volume, , Respectively represent Normalized values ​​of annual historical production and historical sales, , Represent the maximum and minimum values ​​of historical production, , Respectively represent the maximum and minimum values ​​of historical sales;

[0038] Based on the multivariate linear regression method, a regression equation between historical production volume, historical sales volume, policy impact index and green performance index is constructed. The regression equation is expressed as:

[0039] ;

[0040] In the formula Indicates Green Performance Index for the year, Indicates Policy impact index for the year, is the index of the calendar year, , , represents the preset parameter weights in the regression equation, ,and .

[0041] Preferably, the logic of constructing the association model between the green performance index and the historical return data is:

[0042] Based on the support vector regression method, a correlation model between the green performance index and historical return data is constructed. The correlation model is expressed as:

[0043] ;

[0044] In the formula Indicates Years of historical earnings data, Indicates about The kernel function of represents the model error;

[0045] The kernel function uses the radial basis function, which is expressed as:

[0046] ;

[0047] In the formula represents the preset hyperparameters, , represents the L2 norm, represents the preset reference step size, and is a positive integer;

[0048] The mean square error is then used to evaluate the performance of the association model, calculated as:

[0049] ;

[0050] In the formula represents the mean square error, Indicates the number of the calendar year, , Respectively represent the actual collected The historical revenue data of the year and the output of the correlation model Years of historical earnings data;

[0051] The green performance index and historical return data are divided into training set and validation set in a ratio of 4:1. The training set and validation set are used to train and optimize the association model respectively. When the mean square error is less than the preset error threshold, the association model training is considered completed.

[0052] Preferably, the logic for predicting the revenue data for the current year is:

[0053] The time series algorithm adopts an autoregressive integrated moving average model, which is used to perform time series analysis on historical production and historical sales to generate the forecast production and forecast sales for the current year;

[0054] Collect the policy information of the year, substitute the policy information of the year into the text recognition model for processing, and generate the policy impact index of the year;

[0055] Substitute the predicted production volume, predicted sales volume, and policy impact index for the year into the regression equation to generate a predicted performance index, and then substitute the predicted performance index into the association model to generate predicted revenue data.

[0056] Preferably, the forecasted earnings data is compared with the historical earnings data over the years:

[0057] satisfy When the forecast revenue data for the current year is considered to be significantly lower than that of previous years;

[0058] satisfy When the forecast revenue data for the current year is considered to be not much different from previous years;

[0059] satisfy When the forecast revenue data for the current year is considered to have increased significantly compared with previous years;

[0060] In the formula Indicates the preset volatility coefficient, , Represents the forecasted revenue data, It represents the average income data of previous years, and the calculation method is:

[0061] .

[0062] A coupling analysis system between financial technology and green finance, wherein the analysis system is used to execute the above analysis method, including:

[0063] A first data collection module, the first data collection module is used to collect policy information of previous years and policy information of the current year;

[0064] A second data collection module, the second data collection module is used to collect historical output information and historical revenue data of the enterprise;

[0065] A text recognition module, which is used to construct a text recognition model to process policy information over the years and generate a corresponding policy impact index;

[0066] A data processing module, which is used to generate a green performance index of the enterprise on the environment based on the policy impact index and historical output information, and to construct a correlation model between the green performance index and historical revenue data;

[0067] A time series analysis module, which is used to construct a time series algorithm model, perform time series analysis on the historical output information of the enterprise, and generate forecast output information for the current year;

[0068] The comprehensive processing module is used to compare the predicted income data with the historical income data to determine the fluctuation of the predicted income data for the current year.

[0069] Compared with the prior art, the present invention has the following beneficial effects:

[0070] The present invention provides a dynamic and comprehensive coupling analysis method by quantitatively analyzing policy information and corporate performance. By comprehensively considering the impact of multiple factors such as policy impact, corporate production and sales data on corporate revenue, it can effectively measure the performance indicators of enterprises in environmental protection and sustainable development. Compared with existing methods, it improves the objectivity and accuracy of policy evaluation, reduces human subjective factors, and can timely reflect the actual impact of policy changes. At the same time, combined with time series and regression models, it enhances the accuracy and flexibility of revenue forecasting, provides strong technical support for corporate green development decisions, and ensures that enterprises can respond to changes in the external environment in a timely manner, so as to help enterprises better allocate and optimize resources and improve operational efficiency and economic benefits. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Figure 1 It is a schematic diagram of the overall method flow of the present invention;

[0072] Figure 2 It is a schematic diagram of the module structure of the present invention. DETAILED DESCRIPTION

[0073] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with specific embodiments.

[0074] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the present invention shall have the ordinary meanings understood by those of ordinary skill in the field to which the present invention pertains. The "first", "second" and similar terms used in the present invention do not denote any order, quantity or importance, but are only used to distinguish different components. The terms such as "including" or "comprising" mean that the elements or objects appearing before this word cover the elements or objects listed after this word and their equivalents, without excluding other elements or objects. The terms such as "connected" or "coupled" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The terms such as "upper", "lower", "left" and "right" are only used to represent relative positional relationships, and when the absolute position of the object being described changes, the relative positional relationship may also change accordingly.

[0075] Embodiment:

[0076] Please refer to Figure 1 , the present invention provides a technical solution:

[0077] A method for analyzing the coupling of fintech and green finance, the specific steps include:

[0078] S1: Collect the policy information related to the environment over the years, the historical output information and historical revenue data of enterprises. The policy information includes regulatory documents, subsidy documents, and tax documents; the historical output information includes historical production volume and historical sales volume.

[0079] S2: Construct a text recognition model to process the policy information over the years, quantify and score different clauses in the policy information, and generate corresponding policy impact indices.

[0080] The text recognition model adopts a preset natural language processing model. The natural language processing model is an existing technology, and its specific parameters and training methods will not be elaborated here. The logic for quantifying and scoring different clauses in the policy information is as follows:

[0081] Perform document cleaning on different documents in the policy information, remove the header, footer, page numbers, and appendices, only retain the document body, and uniformly convert the document body to the UTF-8 encoding format;

[0082] Segment the document body into phrases, remove the stop words in the document body according to the preset stop word library, and convert different phrases into stem forms. Here, tools such as NLTK and SPACY can be used to process the document body. Stop words refer to common but meaningless words (such as "of", "is", "and"). The preset stop word library can adopt the stop word library自带 by the tool or can be added according to actual needs;

[0083] Calculate the frequency of occurrence of each phrase in the document body, and generate a corresponding importance index based on the frequency of occurrence of each phrase;

[0084] Based on the importance index of each phrase in the document body, a policy relevance index and a corresponding relevance weight are generated, and a policy impact index of the policy information is generated according to the policy relevance index and the relevance weight.

[0085] In this step, the information in the policy document is quantified so that the impact of the policy on the green performance of enterprises can be evaluated more objectively and systematically. Compared with the existing technology, it has the following advantages: 1. By cleaning, segmenting, removing stop words and stemming the text, the information in the policy document can be quantified into measurable indicators, avoiding the bias of subjective judgment, making the policy impact index more objective and repeatable; 2. In the text processing process, irrelevant information such as headers, footers, page numbers and appendices are removed to ensure the high quality of the remaining text, and uniformly converted to UTF-8 encoding to avoid the problems caused by inconsistent encoding, improve the accuracy and consistency of text processing, and thus improve the reliability of subsequent analysis results; 3. By calculating the frequency and importance index of phrases in the document, the most representative and influential parts of the text can be identified, and the key content in the policy document can be highlighted, making the generated policy impact index more representative and explanatory; 4. By calculating the policy relevance index and relevance weight of each document, and combining these indicators to generate the policy impact index, the impact of multiple policy documents can be comprehensively considered to provide an overall policy impact assessment result. Therefore, overall, this step provides a quantitative policy impact indicator, which plays a key role in the construction of green performance index, correlation model analysis, and future benefit forecasting, improves the scientificity and practicality of policy impact assessment, and provides strong technical support for the coupling analysis of financial technology and green finance.

[0086] The frequency of each phrase in the document body is calculated as:

[0087] ;

[0088] In the formula Indicates The phrase in The frequency of occurrence in the body of the document, Indicates The phrase in The number of occurrences in the body of the document, Indicates The total number of phrases in the body of a document. By calculating the frequency and importance of phrases in a document, qualitative information is converted into quantitative indicators, which improves the objectivity and accuracy of policy analysis and avoids human-set bias.

[0089] The importance index of each phrase is calculated as follows:

[0090] ;

[0091] In the formula Indicates The phrase in The importance index in the body of the document, Indicates that it contains The number of documents with the phrase, Represents the total number of documents. The importance index is a variation of the TF-IDF method, which is used to identify phrases that have a greater impact on policies, highlight key policy terms, and make the analysis more targeted and in-depth.

[0092] The policy relevance index for each document body is calculated as:

[0093] ;

[0094] In the formula Indicates The policy relevance index of the document body;

[0095] The relevance weight of each document body is calculated as:

[0096] ;

[0097] The policy impact index of policy information is calculated as follows:

[0098] ;

[0099] In the formula Represents the policy impact index. The calculation of the policy relevance index and relevance weight takes into account the overall impact of each document, providing an indicator for comprehensively evaluating policy impact, thereby reflecting the global effect of the policy. The relevance weight is updated based on the document information each year, ensuring the dynamic nature of the policy impact index and being able to timely reflect the actual impact of policy changes on enterprises.

[0100] S3: Generate the company's green performance index on the environment based on the policy impact index and historical output information, and build a correlation model between the green performance index and historical revenue data.

[0101] The generation logic of the green performance index is:

[0102] The historical output information is normalized and the calculation methods are as follows:

[0103] ;

[0104] ;

[0105] In the formula , Respectively represent Annual historical production volume and historical sales volume, , Respectively represent Normalized values ​​of annual historical production and historical sales, , Represent the maximum and minimum values ​​of historical production, , Respectively represent the maximum and minimum values ​​of historical sales;

[0106] Based on the multivariate linear regression method, a regression equation is constructed between historical production volume, historical sales volume, policy impact index and green performance index. The regression equation is expressed as:

[0107] ;

[0108] In the formula Indicates Green Performance Index for the year, Indicates Policy impact index for the year, is the index of the calendar year, , , represents the preset parameter weights in the regression equation, ,and . The green performance index is an indicator to measure the performance of enterprises in environmental protection and sustainable development. Here, multiple factors such as policy impact, enterprise production and sales data are taken into consideration, so the multivariate linear regression method is used for fitting. The size of the parameter weight is set based on the fact that in the green performance index, the policy impact index corresponding to the policy information on the environment is relatively important, the sales data of the enterprise is second, and the productivity of the enterprise is relatively less important. The specific weight coefficients can be determined through regression analysis or expert opinions to ensure that they can reasonably reflect the importance of each factor.

[0109] The logic of building a correlation model between the green performance index and historical return data is:

[0110] Based on the support vector regression method, a correlation model between the green performance index and historical return data is constructed. The correlation model is expressed as:

[0111] ;

[0112] In the formula Indicates Years of historical earnings data, Indicates about The kernel function of Representing model errors, the support vector regression method uses kernel functions for nonlinear mapping, which can effectively capture complex nonlinear relationships, more accurately reflect the complex relationship between the green performance index and historical return data, and improve the explanatory power of the model.

[0113] The kernel function uses the radial basis function, which is expressed as:

[0114] ;

[0115] In the formula represents the preset hyperparameters, , represents the L2 norm, represents the preset reference step size, and is a positive integer. , represents the input vector. Since the kernel function only considers the green performance index, the input vector only contains The distance between the input vectors determines the value of the kernel function. Points that are closer will have larger kernel function values, indicating that they are more similar in the high-dimensional space; points that are farther away will have smaller kernel function values, indicating that they are more different in the high-dimensional space. This kernel function can be regarded as The kernel function is a Gaussian distribution centered on , and its shape is controlled by a hyperparameter. When the hyperparameter is large, the shape of the Gaussian distribution becomes narrow and sharp; when the hyperparameter is small, the shape of the Gaussian distribution becomes wide and flat. This geometric property enables the kernel function to effectively handle nonlinear relationships and better reflect the relationship between the green performance index and historical return data.

[0116] The mean square error is then used to evaluate the performance of the association model, calculated as:

[0117] ;

[0118] In the formula represents the mean square error, Indicates the number of the calendar year, , Respectively represent the actual collected The historical revenue data of the year and the output of the correlation model Years of historical earnings data;

[0119] The green performance index and historical return data are divided into training set and validation set in a ratio of 4:1. The training set and validation set are used to train and optimize the association model respectively. When the mean square error is less than the preset error threshold, the association model training is considered completed.

[0120] In this step, by constructing a correlation model between the green performance index and historical earnings data, we can accurately capture complex relationships and improve forecast accuracy, helping companies understand the specific impact of green performance on earnings, providing a scientific basis for decision-making optimization, and facilitating longer-term strategic planning for companies.

[0121] S4: Build a time series algorithm model, conduct time series analysis on the company's historical output information, and generate forecast output information for the current year. The forecast output information includes forecast production volume and forecast sales volume.

[0122] S5: Collect the policy information related to the environment in the current year, generate a forecast performance index based on the policy information and forecast output information of the current year, and then forecast the revenue data of the current year based on the forecast performance index.

[0123] The logic for predicting the revenue data for the current year is:

[0124] The time series algorithm uses an autoregressive integrated moving average model, which is used to perform time series analysis on historical production and historical sales to generate the forecast production and forecast sales for the current year;

[0125] Collect the policy information of the year, substitute the policy information of the year into the text recognition model for processing, and generate the policy impact index of the year;

[0126] Substitute the predicted production volume, predicted sales volume, and policy impact index for the year into the regression equation to generate a predicted performance index, and then substitute the predicted performance index into the association model to generate predicted revenue data.

[0127] In this step, by utilizing time series analysis and regression models, we can consider the impact of historical data and current policies at the same time, providing an all-round, multi-dimensional forecasting model, which improves the comprehensiveness and accuracy of the forecasting results, and can help companies make more scientific and reasonable strategic decisions, enhance their market competitiveness and sustainable development capabilities. At the same time, it can also respond and adjust to policy changes in real time, improve the flexibility and adaptability of the model, and ensure that companies can respond to changes in the external environment in a timely manner, so as to help companies better allocate and optimize resources and improve operational efficiency and economic benefits.

[0128] Compare the forecasted earnings data with the historical earnings data over the years:

[0129] satisfy When the forecast revenue data for the current year is considered to be significantly lower than that of previous years;

[0130] satisfy When the forecast revenue data for the current year is considered to be not much different from previous years;

[0131] satisfy When the forecast revenue data for the current year is considered to have increased significantly compared with previous years;

[0132] In the formula Indicates the preset volatility coefficient, , Represents the forecasted revenue data, It represents the average income data of previous years, and the calculation method is:

[0133] .

[0134] In summary, the present invention provides a dynamic and comprehensive coupling analysis method by quantitatively analyzing policy information and corporate performance. By comprehensively considering the impact of multiple factors such as policy impact, corporate production and sales data on corporate revenue, it can effectively measure the performance indicators of enterprises in environmental protection and sustainable development. Compared with existing methods, the objectivity and accuracy of policy evaluation are improved, and the actual impact of policy changes can be reflected in a timely manner. At the same time, combined with time series and regression models, the accuracy and flexibility of revenue forecasts are enhanced, providing strong technical support for corporate green development decisions, ensuring that companies can respond to changes in the external environment in a timely manner, so as to help companies better allocate and optimize resources and improve operational efficiency and economic benefits.

[0135] See also Figure 2 The present invention also provides a financial technology and green finance coupling analysis system for executing the above analysis method, including:

[0136] The first data collection module is used to collect policy information of previous years and policy information of the current year;

[0137] The second data collection module is used to collect the historical output information and historical revenue data of the enterprise;

[0138] The text recognition module is used to build a text recognition model to process policy information over the years and generate the corresponding policy impact index;

[0139] Data processing module: The data processing module is used to generate the green performance index of the enterprise on the environment based on the policy impact index and historical output information, and to build a correlation model between the green performance index and historical revenue data;

[0140] Time series analysis module: The time series analysis module is used to build a time series algorithm model, conduct time series analysis on the company's historical output information, and generate the forecast output information for the current year;

[0141] The comprehensive processing module is used to compare the predicted income data with the historical income data to determine the fluctuation of the predicted income data for the current year.

[0142] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0143] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product. Those skilled in the art may appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein may be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed by hardware or software methods depends on the specific application and design constraints of the technical solution.

[0144] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, and may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0145] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. The coupling analysis method between financial technology and green finance is characterized by: The specific steps include: S1: Collect policy information related to the environment over the years, as well as the historical output information and historical revenue data of the enterprise. The historical output information includes historical production volume and historical sales volume, and the predicted output information includes predicted production volume and predicted sales volume; S2: Construct a text recognition model to process policy information of previous years, quantify and score different clauses in the policy information, and generate the policy impact index of the corresponding year. The text recognition model uses a preset natural language processing model, and the logic of quantifying and scoring different clauses in the policy information is as follows: Clean the different documents in the policy information and only keep the document body; Split the document body into phrases; Calculate the frequency of occurrence of each phrase in the document body, and generate a corresponding importance index based on the frequency of occurrence of each phrase; Based on the importance index of each phrase in the document body, a policy relevance index and a corresponding relevance weight are generated, and a policy impact index of the policy information is generated according to the policy relevance index and the relevance weight; The policy relevance index for each document body is calculated as: ； In the formula Indicates The policy relevance index of the document body, Indicates The total number of phrases in the body of the document, Indicates The phrase in Importance index in the body of the document; The relevance weight of each document body is calculated as: ； The policy impact index of policy information is calculated as follows: ； In the formula represents the policy impact index, No. The relevance weight of the document body, Indicates the total number of documents; S3: Generate the green performance index of enterprises on the environment based on the policy impact index and historical output information, and build a correlation model between the green performance index and historical revenue data; The generation logic of the green performance index is: Normalize historical output information; Based on the multivariate linear regression method, a regression equation between historical production volume, historical sales volume, policy impact index and green performance index was constructed; The logic of building a correlation model between the green performance index and historical return data is: Based on the support vector regression method, a correlation model between the green performance index and historical return data is constructed. The correlation model is expressed as: ； In the formula Indicates Years of historical earnings data, Indicates about The kernel function of Indicates Green Performance Index for the year, represents the model error, An index of the calendar year; The kernel function uses the radial basis function, which is expressed as: ； In the formula represents the preset hyperparameters, , represents the L2 norm, represents the preset reference step size, and is a positive integer; The mean square error is then used to evaluate the performance of the association model; S4: Build a time series algorithm model to conduct time series analysis on the company's historical output information and generate the forecast output information for the current year; S5: Collect the policy information related to the environment in the current year, generate a forecast performance index based on the policy information and forecast output information of the current year, and then forecast the revenue data of the current year based on the forecast performance index.

2. The method for analyzing the coupling between financial technology and green finance according to claim 1 is characterized by: The calculation methods for normalizing historical output information are: ； ； In the formula , Respectively represent Annual historical production volume and historical sales volume, , Respectively represent Normalized values ​​of annual historical production and historical sales, , Represent the maximum and minimum values ​​of historical production respectively, , Respectively represent the maximum and minimum values ​​of historical sales; The regression equation between historical production volume, historical sales volume, policy impact index and green performance index based on the multivariate linear regression method is expressed as: ； In the formula Indicates Policy impact index for the year, , , represents the preset parameter weights in the regression equation, ,and .

3. The method for analyzing the coupling between financial technology and green finance according to claim 2 is characterized by: The logic for predicting the revenue data for the current year is: The time series algorithm adopts an autoregressive integrated moving average model, which is used to perform time series analysis on historical production and historical sales to generate the forecast production and forecast sales for the current year; Collect the policy information of the year, substitute the policy information of the year into the text recognition model for processing, and generate the policy impact index of the year; Substitute the predicted production volume, predicted sales volume, and policy impact index for the year into the regression equation to generate a predicted performance index, and then substitute the predicted performance index into the association model to generate predicted revenue data.

4. The method for analyzing the coupling between financial technology and green finance according to claim 3 is characterized by: Compare the forecasted earnings data with the historical earnings data over the years: satisfy When the forecast revenue data for the current year is considered to be significantly lower than that of previous years; satisfy When the forecast revenue data for the current year is considered to be not much different from previous years; satisfy When the forecast revenue data for the current year is considered to have increased significantly compared with previous years; In the formula Indicates the preset volatility coefficient, , Represents the forecasted revenue data, It represents the average income data of previous years, which is calculated as follows: ； In the formula Indicates the total number of calendar years.

5. The coupling analysis system of financial technology and green finance is characterized by: The analysis system is used to perform the analysis method according to any one of claims 1 to 4, comprising: A first data collection module, the first data collection module is used to collect policy information of previous years and policy information of the current year; A second data collection module, the second data collection module is used to collect historical output information and historical revenue data of the enterprise; A text recognition module, which is used to construct a text recognition model to process policy information over the years and generate a corresponding policy impact index; A data processing module, which is used to generate a green performance index of the enterprise on the environment based on the policy impact index and historical output information, and to construct a correlation model between the green performance index and historical revenue data; A time series analysis module, which is used to construct a time series algorithm model, perform time series analysis on the historical output information of the enterprise, and generate forecast output information for the current year; The comprehensive processing module is used to compare the predicted income data with the historical income data to determine the fluctuation of the predicted income data for the current year.

Images