Carbon asset portfolio weight recommendation method and system

Abstract

The invention provides a carbon asset portfolio weight recommendation method and system, and relates to the technical field of big data processing, and the method comprises the steps: receiving multiple pieces of target carbon asset information sent by a client; obtaining carbon asset data corresponding to the carbon asset names in a carbon asset database, and calculating dynamic correlation coefficients of each target carbon asset in a historical investment period to obtain a covariance matrix; calculating according to the covariance matrix to obtain a combined volatility and an absolute contribution value of each target carbon asset; constructing a recommendation model based on the combination volatility and the absolute contribution value, and continuously performing iterative optimization to obtain an investment combination weight vector; and sending the portfolio weight vectors to the client, so that the client displays the multiple pieces of target carbon asset information and the corresponding portfolio weight vectors at the same time. According to the method, the problems that the current investment portfolio weight recommendation is unreasonable, non-objective and low in efficiency are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of big data processing, and in particular to a method and system for dynamically recommending carbon asset investment portfolio weights based on Bayesian optimization. Background Art

[0002] As the global climate change problem intensifies and the pace of carbon emission reduction accelerates, it is very important to establish and improve the carbon asset market trading mechanism to promote greenhouse gas emission reduction and promote the green and low-carbon transformation of the economy.

[0003] As an important means of regulating carbon emissions, the carbon asset market is seeing an increasing scale of transactions and complexity. Traditional carbon asset investment portfolios are mainly conducted by professionals who conduct quantitative analysis of historical data and recommend investment portfolio weights to users one-on-one. Although this method can balance the risks of various carbon assets in the carbon asset investment portfolio to a certain extent, it ignores the changing laws of the dynamic correlation between different carbon assets, resulting in the inability to quantify the linkage risks between different carbon assets in multiple carbon asset configurations. In the face of a complex carbon market environment, not only is the carbon asset investment portfolio weight recommendation unreasonable and subjective, and prone to falling into local optimality, but it also requires high labor costs, and the investment portfolio weight recommendation is highly subjective and inefficient. Summary of the invention

[0004] The present invention provides a carbon asset investment portfolio weight recommendation method and system, which are used to solve the problem that current investment portfolio weight recommendation is unreasonable, subjective and inefficient.

[0005] In one aspect, the present invention provides a method for recommending carbon asset investment portfolio weights, the method being applied to a recommendation system, the recommendation system comprising a client, the recommendation system interacting with a user through the client; the method comprising: Receiving a plurality of target carbon asset information sent by the client; the target carbon asset information includes a carbon asset name and an investment period; Acquire carbon asset data corresponding to the carbon asset name in the carbon asset database, calculate the dynamic correlation coefficients of each target carbon asset in the historical investment cycle, and obtain a covariance matrix; Calculate the combined volatility and the absolute contribution value of each target carbon asset according to the covariance matrix; Building a recommendation model based on the portfolio volatility and the absolute contribution value, and continuously performing iterative optimization to obtain an investment portfolio weight vector; The investment portfolio weight vector is sent to the client, so that the client simultaneously displays a plurality of target carbon asset information and the corresponding investment portfolio weight vector.

[0006] Furthermore, the dynamic correlation coefficient is obtained by: An unstandardized correlation matrix is ​​obtained by using time series analysis, and the unstandardized correlation matrix is ​​normalized to obtain a dynamic correlation coefficient.

[0007] Furthermore, the calculation formula of the dynamic correlation coefficient is:

[0008] in, Indicates that in the historical investment cycle Internal Carbon Assets and carbon assets The dynamic correlation coefficient between Indicates that in the historical investment cycle Carbon assets Relative to carbon assets Fluctuation value of Indicates that in the historical investment cycle Carbon assets Fluctuation value of Indicates that in the historical investment cycle Carbon assets The fluctuation value of .

[0009] Furthermore, the covariance matrix The calculation method for each element in is: ; in, Represents carbon assets in the historical investment cycle and carbon assets The dynamic correlation coefficient between Represents carbon assets in the historical investment cycle volatility; Represents carbon assets in the historical investment cycle volatility.

[0010] Furthermore, the calculation formula of the combined volatility is: ; in, represents the portfolio weight vector, , represents the amount of carbon assets; represents the covariance matrix.

[0011] Furthermore, the calculation formula for the absolute contribution value of each target carbon asset is: ; in, Represents carbon assets The weight vector of Represents carbon assets marginal contribution to the carbon asset portfolio; Represents carbon assets The weight, represents the covariance matrix, represents the portfolio weight vector; represents the portfolio volatility.

[0012] Furthermore, the construction of the recommendation model includes: Establishing an objective function based on the portfolio volatility and the absolute contribution value to make the contribution value of each carbon asset in the portfolio equal; Get the initial weight vector, use the Gaussian process to fit the posterior distribution of the objective function, and get the recommendation model; In each iteration, the next portfolio weight vector is selected and updated continuously until the objective function value changes less than the set threshold, and the iteration is stopped and the portfolio weight vector is output.

[0013] Furthermore, the recommendation model is: ; in, represents the mean function of the objective function; represents the current optimal solution; represents the cumulative distribution function of the standard normal distribution; represents the uncertainty of the Gaussian process; Represents the probability density function of the standard normal distribution.

[0014] Furthermore, the objective function for: ; in, represents the portfolio weight vector, ; represents the amount of carbon assets; Represents carbon assets The absolute contribution value of represents the portfolio volatility.

[0015] On the other hand, the present invention also provides a carbon asset investment portfolio weight recommendation system, the recommendation system comprising a client and a first module, a second module and a third module connected in sequence; The client is used to interact with the user, receive multiple target carbon asset information input by the user and send it to the first module, and receive the investment portfolio weight vector from the third module, and display multiple target carbon asset information and the corresponding investment portfolio weight vector; the target carbon asset information includes the carbon asset name and the focus period; The first module is used to receive information of multiple target carbon assets, obtain carbon asset data corresponding to the carbon asset name from the carbon asset database, calculate the dynamic correlation coefficients of each target carbon asset in the historical investment cycle, and obtain a covariance matrix; The second module is used to obtain the covariance matrix, and calculate the combined volatility and the absolute contribution value of each target carbon asset according to the covariance matrix; The third module is used to obtain the combination volatility and the absolute contribution value, establish a recommendation model, continuously perform iterative optimization, obtain an investment portfolio weight vector and send it to the client.

[0016] In general, the present invention provides a method and system for recommending carbon asset investment portfolio weights. Compared with the prior art, the technical solution conceived by the present invention can achieve the following beneficial effects: The present invention establishes a recommendation model based on the target carbon asset information provided by the user by calculating the dynamic correlation coefficients of each target carbon asset within the investment cycle, quantifies the correlation between different carbon assets, and obtains the investment portfolio weight vector through continuous iterative optimization. This not only improves the intelligence and automation of investment portfolio weight optimization, increases the rationality and objectivity of weight recommendation, but also reduces manual intervention, adapts to the complex and changeable carbon market environment, and improves the efficiency of carbon market investment portfolio management. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1 It is a method flow diagram of a carbon asset investment portfolio weight recommendation method and system provided by the present invention; Figure 2 It is a system structure diagram of a carbon asset investment portfolio weight recommendation method and system provided by the present invention. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings and embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work belong to the scope of protection of the present invention.

[0020] It should be noted that, in the description of the embodiments of the present invention, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a method, step or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such method, step or device. In the absence of further restrictions, the elements defined by the sentence "include a ..." do not exclude the existence of other identical elements in the method, step or device including the elements.

[0021] In order to solve the problem that current investment portfolio weight recommendation is unreasonable, subjective and inefficient, the present invention provides a carbon asset investment portfolio weight recommendation method, which is applied to a recommendation system. The recommendation system includes a client, and the recommendation system interacts with the user through the client. Specifically, Figure 1 As shown, the method includes: S101: receiving multiple target carbon asset information sent by a client.

[0022] Among them, the target carbon asset information includes the carbon asset name and investment period.

[0023] It should be noted that carbon assets mainly include three categories based on trading mechanisms and emission reduction sources: carbon quotas, carbon credits and carbon financial derivatives.

[0024] Carbon quotas are greenhouse gas emission quotas, such as the European Union Carbon Allowances (EUA) and China Certified Emission Reductions (CCER).

[0025] Carbon credits are actual carbon emission reductions achieved through specific projects, such as Clean Development Mechanism Emission Reductions (CERs) and Voluntary Emission Reductions (VERs).

[0026] Carbon financial derivatives are derivatives based on carbon quotas or carbon credits, including carbon futures, carbon options and carbon swaps, such as EUA futures and CCER futures.

[0027] The multiple target carbon assets can be the same type of carbon assets or different types of carbon assets; the investment period is the investment period preset by the user.

[0028] S102: Obtain carbon asset data corresponding to the carbon asset name in the carbon asset database, calculate the dynamic correlation coefficients of each target carbon asset in the historical investment cycle, and obtain a covariance matrix.

[0029] The carbon asset database is a carbon asset data set that uses the carbon asset name as a data package. Users can directly select the name of the target carbon asset to be invested in from the carbon asset database and then obtain all historical related data of the target carbon asset.

[0030] It should be noted that carbon asset data includes carbon asset names and historical asset data. For example, a data table can be established with the carbon asset name as the table name, time as the column coordinate, and historical asset data as the horizontal coordinate. Among them, historical asset data can include: price, code, sector, industry, company, investment risk level, total amount, transaction amount, remaining amount, turnover rate, yield, volatility and other data related to the carbon asset.

[0031] The historical investment cycle is greater than or equal to the investment cycle set by the user. The historical investment cycle is obtained by taking the current user-set time as the base point and tracing back to the past investment cycle greater than or equal to the user-set investment cycle to obtain the historical investment cycle.

[0032] For example, the current user sets the time to January 1, 2025, and the investment cycle is set to one year, that is, January 1, 2025 to January 1, 2026. The historical investment cycle is greater than, which can be January 1, 2024 to January 1, 2025, or December 1, 2023 to January 1, 2025. The carbon asset data from January 1, 2024 to January 1, 2025 or December 1, 2023 to January 1, 2025 is obtained to calculate the dynamic correlation coefficient of each target carbon asset within the historical investment cycle.

[0033] Since there is a dynamic correlation between different carbon assets, it will have an important impact on the investment portfolio. In order to quantify the dynamic correlation between carbon assets, the DCC-GARCH model is used for measurement. As an embodiment, the method for obtaining the dynamic correlation coefficient is: using time series analysis to obtain an unstandardized correlation matrix, normalizing the unstandardized correlation matrix, and obtaining the dynamic correlation coefficient. Specifically, the unstandardized correlation matrix for: ; in, represents the coefficient of impact of historical investment cycle shocks on dynamic correlation; represents the persistence parameter of the dynamic correlation due to historical investment cycle shocks; represents the historical investment cycle mean correlation matrix, represents the volatility residual of the previous historical investment cycle, Represents the mean of the volatility residuals of the previous historical investment cycle.

[0034] In order to obtain a standardized correlation matrix, the unstandardized correlation matrix needs to be normalized and the dynamic correlation coefficient calculated.

[0035] As an embodiment, the calculation formula of the dynamic correlation coefficient is:

[0036] in, Indicates that in the historical investment cycle Internal Carbon Assets and carbon assets The dynamic correlation coefficient between Indicates that in the historical investment cycle Carbon assets Relative to carbon assets The fluctuation value, that is, the unstandardized correlation matrix The OK The element value of the column; Indicates that in the historical investment cycle Carbon assets The fluctuation value, that is, the unstandardized correlation matrix The OK The element value of the column; Indicates that in the historical investment cycle Carbon assets The fluctuation value, that is, the unstandardized correlation matrix The OK The element value of the column.

[0037] As an example, the covariance matrix The calculation method for each element in is: ; in, Represents carbon assets in the historical investment cycle and carbon assets The dynamic correlation coefficient between Represents carbon assets in the historical investment cycle volatility; Represents carbon assets in the historical investment cycle volatility.

[0038] It should be noted that volatility is a key indicator to measure the drastic degree of asset changes. The volatility of carbon assets usually has clustering and non-normal distribution characteristics, which can be directly obtained from the carbon asset database; the time series analysis model (GARCH) can also be used to describe the volatility of carbon assets. As an example, the volatility calculation method can be: ; in, Indicates the historical investment cycle of carbon assets Volatility within It represents the long-term average of volatility and indicates the basic volatility level; Indicates that in the historical investment cycle The degree to which the impact of the internal carbon asset shock affects the volatility at the current moment; Indicates the degree of continuous impact of the volatility of the previous historical investment cycle on the volatility of the current cycle; Represents the volatility residual of the previous historical investment cycle.

[0039] S103: Calculate the portfolio volatility and the absolute contribution value of each target carbon asset based on the covariance matrix.

[0040] As an example, the calculation formula of the combined volatility is: ; in, represents the portfolio weight vector, , represents the amount of carbon assets; represents the covariance matrix.

[0041] As an example, the calculation formula for the absolute contribution value of each target carbon asset is: ; in, Represents carbon assets The weight vector of Represents carbon assets marginal contribution to the carbon asset portfolio; Represents carbon assets The weight, represents the covariance matrix, represents the portfolio weight vector; represents the portfolio volatility.

[0042] S104: Build a recommendation model based on the portfolio volatility and absolute contribution value, and continuously perform iterative optimization to obtain the investment portfolio weight vector.

[0043] As an embodiment, the construction of the recommendation model includes: S401: Establish an objective function based on the portfolio volatility and absolute contribution value to make the contribution value of each carbon asset in the portfolio equal.

[0044] Furthermore, the objective function Can be: ; in, represents the portfolio weight vector, ; represents the amount of carbon assets; Represents carbon assets The absolute contribution value of represents the portfolio volatility.

[0045] S402: Obtain an initial weight vector, and use a Gaussian process to fit the posterior distribution of the objective function to obtain a recommendation model.

[0046] It should be noted that the initial weight vector can be initially set by the user or calculated based on the number of carbon assets. For example, the initial weight vector can be , .

[0047] Furthermore, the posterior distribution of the objective function fitted using the Gaussian process can be: ;in, represents the mean function of the objective function; Represents the covariance function of the objective function.

[0048] The recommendation model selects the next portfolio weight in each iteration by maximizing the expected improvement, that is: .in, Indicates the current optimal value; Represents the current solved portfolio weight vector The corresponding objective function value.

[0049] Therefore, as an embodiment, the recommendation model may be: ; in, represents the mean function of the objective function; represents the current optimal solution; represents the cumulative distribution function of the standard normal distribution; represents the uncertainty of the Gaussian process; Represents the probability density function of the standard normal distribution.

[0050] S403: Select the next investment portfolio weight vector in each iteration, and continuously iterate and update until the objective function value changes less than a set threshold, then stop the iteration and output the investment portfolio weight vector.

[0051] It should be noted that in the iterative update process, the gradient descent method can be used to iteratively update the portfolio weight vector, specifically: ; in, Represents carbon assets In the The portfolio weight vector for the iteration; represents the learning rate; Represents the objective function.

[0052] S105: Sending the investment portfolio weight vector to the client, so that the client can simultaneously display multiple target carbon asset information and corresponding investment portfolio weight vectors.

[0053] In a second aspect, the present invention also provides a carbon asset investment portfolio weight recommendation system, such as Figure 2 As shown, the recommendation system includes a client and a first module, a second module and a third module connected in sequence.

[0054] The client is used to interact with the user, receive multiple target carbon asset information input by the user and send it to the first module, and receive the investment portfolio weight vector from the third module, and display multiple target carbon asset information and corresponding investment portfolio weight vector; the target carbon asset information includes the carbon asset name and the focus period; The first module is used to receive information of multiple target carbon assets, obtain carbon asset data corresponding to the carbon asset name from the carbon asset database, calculate the dynamic correlation coefficients of each target carbon asset in the historical investment cycle, and obtain the covariance matrix; The second module is used to obtain the covariance matrix, and calculate the portfolio volatility and the absolute contribution value of each target carbon asset based on the covariance matrix; The third module is used to obtain the portfolio volatility and absolute contribution value, establish a recommendation model, continuously perform iterative optimization, obtain the investment portfolio weight vector and send it to the client.

[0055] The technical features of the system are consistent with the technical features of the above method and will not be described in detail here.

[0056] It should be noted that, for the above-mentioned various embodiments, for the sake of simplicity of description, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.

[0057] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0058] In the several embodiments provided in this application, it should be understood that the disclosed method or system can be implemented in other ways. For example, the above-described embodiments are only schematic, and the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed.

[0059] 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, that is, they 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.

[0060] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0061] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or the whole or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a memory, including a number of instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application.

[0062] Those skilled in the art will appreciate that all or part of the various circuits in the above embodiments may be completed by entering a program to instruct related hardware, and the program may be stored in a computer-readable memory, which may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0063] The above is only an exemplary embodiment of the present disclosure, and the scope of the present disclosure cannot be limited thereto. That is, any equivalent changes and modifications made according to the teachings of the present disclosure are still within the scope of the present disclosure. After considering the specification and practicing the disclosure here, those skilled in the art will easily think of the implementation scheme of the present disclosure. This application is intended to cover any modification, use or adaptation of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary technical means in the technical field not recorded in the present disclosure. The description and examples are regarded as exemplary only, and the scope and spirit of the present disclosure are defined by the claims.

[0064] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0065] It will be easily understood by those skilled in the art that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A carbon asset investment portfolio weight recommendation method, characterized in that: The method is applied to a recommendation system, the recommendation system includes a client, and the recommendation system interacts with a user through the client; the method includes: Receiving a plurality of target carbon asset information sent by the client; the target carbon asset information includes a carbon asset name and an investment period; Acquire carbon asset data corresponding to the carbon asset name in the carbon asset database, calculate the dynamic correlation coefficients of each target carbon asset in the historical investment cycle, and obtain a covariance matrix; Calculate the combined volatility and the absolute contribution value of each target carbon asset according to the covariance matrix; Building a recommendation model based on the portfolio volatility and the absolute contribution value, and continuously performing iterative optimization to obtain an investment portfolio weight vector; The investment portfolio weight vector is sent to the client, so that the client simultaneously displays a plurality of target carbon asset information and the corresponding investment portfolio weight vector.

2. A carbon asset investment portfolio weight recommendation method according to claim 1, characterized in that: The method for obtaining the dynamic correlation coefficient is: An unstandardized correlation matrix is ​​obtained by using time series analysis, and the unstandardized correlation matrix is ​​normalized to obtain a dynamic correlation coefficient.

3. A carbon asset investment portfolio weight recommendation method according to claim 2, characterized in that: The calculation formula of the dynamic correlation coefficient is: in, Indicates that in the historical investment cycle Internal Carbon Assets and carbon assets The dynamic correlation coefficient between Indicates that in the historical investment cycle Carbon assets Relative to carbon assets Fluctuation value of Indicates that in the historical investment cycle Carbon assets Fluctuation value of Indicates that in the historical investment cycle Carbon assets The fluctuation value of .

4. A carbon asset investment portfolio weight recommendation method according to claim 3, characterized in that: The covariance matrix The calculation method for each element in is: ； in, Represents carbon assets in the historical investment cycle and carbon assets The dynamic correlation coefficient between Represents carbon assets in the historical investment cycle volatility; Represents carbon assets in the historical investment cycle volatility.

5. A carbon asset investment portfolio weight recommendation method according to claim 1, characterized in that: The calculation formula of the combined volatility is: ； in, represents the portfolio weight vector, , represents the amount of carbon assets; represents the covariance matrix.

6. A carbon asset investment portfolio weight recommendation method according to claim 1, characterized in that: The calculation formula for the absolute contribution value of each target carbon asset is: ； in, Represents carbon assets The weight vector of Represents carbon assets marginal contribution to the carbon asset portfolio; Represents carbon assets The weight, represents the covariance matrix, represents the portfolio weight vector; represents the portfolio volatility.

7. A carbon asset investment portfolio weight recommendation method according to claim 1, characterized in that: The construction of the recommendation model includes: Establishing an objective function based on the portfolio volatility and the absolute contribution value to make the contribution value of each carbon asset in the portfolio equal; Get the initial weight vector, use the Gaussian process to fit the posterior distribution of the objective function, and get the recommendation model; In each iteration, the next portfolio weight vector is selected and updated continuously until the objective function value changes less than the set threshold, and the iteration is stopped and the portfolio weight vector is output.

8. A carbon asset investment portfolio weight recommendation method according to claim 7, characterized in that: The recommended model is: ； in, represents the mean function of the objective function; represents the current optimal solution; represents the cumulative distribution function of the standard normal distribution; represents the uncertainty of the Gaussian process; Represents the probability density function of the standard normal distribution.

9. A carbon asset investment portfolio weight recommendation method according to claim 8, characterized in that: The objective function for: ； in, represents the portfolio weight vector, ; represents the amount of carbon assets; Represents carbon assets The absolute contribution value of represents the portfolio volatility.

10. A carbon asset investment portfolio weight recommendation system, characterized in that: The recommendation system includes a client and a first module, a second module and a third module connected in sequence; The client is used to interact with the user, receive multiple target carbon asset information input by the user and send it to the first module, and receive the investment portfolio weight vector from the third module, and display multiple target carbon asset information and the corresponding investment portfolio weight vector; the target carbon asset information includes the carbon asset name and the focus period; The first module is used to receive information of multiple target carbon assets, obtain carbon asset data corresponding to the carbon asset name from the carbon asset database, calculate the dynamic correlation coefficients of each target carbon asset in the historical investment cycle, and obtain a covariance matrix; The second module is used to obtain the covariance matrix, and calculate the combined volatility and the absolute contribution value of each target carbon asset according to the covariance matrix; The third module is used to obtain the combination volatility and the absolute contribution value, establish a recommendation model, continuously perform iterative optimization, obtain an investment portfolio weight vector and send it to the client.

Images