CN115482041A - Bond valuation method, bond valuation model training method and device - Google Patents

Bond valuation method, bond valuation model training method and device Download PDF

Info

Publication number
CN115482041A
CN115482041A CN202211177561.XA CN202211177561A CN115482041A CN 115482041 A CN115482041 A CN 115482041A CN 202211177561 A CN202211177561 A CN 202211177561A CN 115482041 A CN115482041 A CN 115482041A
Authority
CN
China
Prior art keywords
bond
price
evaluated
valuation
bonds
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202211177561.XA
Other languages
Chinese (zh)
Inventor
徐阆平
陈风
郭立帆
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Alibaba Cloud Computing Ltd
Original Assignee
Alibaba Cloud Computing Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Alibaba Cloud Computing Ltd filed Critical Alibaba Cloud Computing Ltd
Priority to CN202211177561.XA priority Critical patent/CN115482041A/en
Publication of CN115482041A publication Critical patent/CN115482041A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0283—Price estimation or determination
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
    • G06Q40/04—Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
    • G06Q40/06—Asset management; Financial planning or analysis

Landscapes

  • Business, Economics & Management (AREA)
  • Engineering & Computer Science (AREA)
  • Accounting & Taxation (AREA)
  • Finance (AREA)
  • Development Economics (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Strategic Management (AREA)
  • General Physics & Mathematics (AREA)
  • Marketing (AREA)
  • Economics (AREA)
  • General Business, Economics & Management (AREA)
  • Technology Law (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Game Theory and Decision Science (AREA)
  • Operations Research (AREA)
  • Evolutionary Computation (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Human Resources & Organizations (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The application provides a bond valuation method, a bond valuation model training method and a device, which relate to the field of cloud computing and comprise the following steps: determining the sparse characteristics of the bonds to be evaluated based on the historical prices of the bonds to be evaluated and the real-time bargaining prices of the bonds of the same type of the bonds to be evaluated; inputting the sparse characteristics into a bond valuation model, and obtaining a future transaction price valuation of the bond to be valued based on an output value of the bond valuation model; and displaying the price valuation of the future deal through a bond price display page. In the embodiment, the evaluation model of the bond is trained by using the historical bargaining price of the bond to be evaluated and the historical bargaining price of the bond of the same type, the bond is evaluated by using the evaluation model of the bond, and the evaluation of the bond is carried out based on the historical price of the bond to be evaluated and the real-time bargaining price of the bond of the same type of the bond to be evaluated, so that the real-time evaluation can be realized, and the evaluation result is more accurate.

Description

Bond valuation method, bond valuation model training method and device
Technical Field
The application relates to the technical field of cloud computing, in particular to a bond valuation method, a bond valuation model training method and a bond valuation model training device.
Background
Evaluating a bond refers to the process of estimating the future transaction price of the bond. Generally, when the bond is estimated, the real-time transaction price is used for estimating. However, for some bonds, the deal data is sparse, e.g., credit bonds. At present, in a credit and debt trading market, some credit and debts cannot be traded for several days, so that the real-time trading price cannot be directly used for valuation of the credit and debt, but the real-time valuation can help traders to quote more conveniently.
The existing bond evaluation method can only evaluate bonds with sparse transaction data by taking days as units, and can not realize real-time evaluation and inaccurate evaluation results after closing a balance every day.
Disclosure of Invention
The embodiment of the application provides a bond valuation method, a bond valuation model training method and a bond valuation model training device, so that real-time valuation of bonds is achieved, and valuation results are more accurate.
In a first aspect, an embodiment of the present application provides a bond valuation method, where the method includes:
determining the sparse characteristics of the bonds to be evaluated based on the historical prices of the bonds to be evaluated and the real-time bargaining prices of the bonds of the same type of the bonds to be evaluated;
inputting the sparse characteristics into a bond valuation model, and obtaining a future transaction price valuation of the bond to be valued based on an output value of the bond valuation model;
displaying the future transaction price valuation through a bond price display page;
the bond valuation model is obtained by training by utilizing the historical trading price of the bond to be valued and the historical trading price of the bond of the same type.
In a second aspect, an embodiment of the present application provides a method for training a bond valuation model, where the method includes:
acquiring a training sample set, wherein the training sample set comprises training samples and sample labels, the training samples comprise sparse characteristics obtained based on historical transaction prices of bonds to be evaluated and historical transaction prices of bonds of the same type, and the sample labels comprise the difference between the historical transaction prices of the bonds to be evaluated and the historical transaction prices of the bonds of the same type;
and training the neural network model by utilizing the training sample set to obtain a bond valuation model of the bond to be valued.
In a third aspect, an embodiment of the present application provides a bond valuation apparatus, including:
the characteristic determination module is used for determining the sparse characteristic of the bond to be evaluated based on the historical price of the bond to be evaluated and the real-time transaction price of the bond of the same type of the bond to be evaluated;
the price determination module is used for inputting the sparse characteristics into the bond valuation model and obtaining a future transaction price valuation of the bond to be valued based on the output value of the bond valuation model;
the price display module is used for displaying the price valuation of the future transaction through a bond price display page;
the bond valuation model is obtained by training by utilizing the historical trading price of the bond to be valued and the historical trading price of the bond of the same type.
In a fourth aspect, an embodiment of the present application provides a bond valuation model training device, where the device includes:
the system comprises a sample acquisition module, a comparison module and a comparison module, wherein the sample acquisition module is used for acquiring a training sample set, the training sample set comprises training samples and sample labels, the training samples comprise sparse features obtained based on historical transaction prices of bonds to be evaluated and historical transaction prices of bonds of the same type, and the sample labels comprise the difference between the historical transaction prices of the bonds to be evaluated and the historical transaction prices of the bonds of the same type;
and the model training module is used for training the neural network model by utilizing the training sample set to obtain a bond evaluation model of the bond to be evaluated.
In a fifth aspect, an embodiment of the present application provides an electronic device, including a memory, a processor, and a computer program stored on the memory, where the processor implements the method of any one of the above when executing the computer program.
Compared with the prior art, the method has the following advantages:
the embodiment of the application provides a bond valuation method, a bond valuation model training method and a device, and the sparse characteristics of bonds to be valued are determined based on the historical prices of the bonds to be valued and the real-time bargaining prices of bonds of the same type of the bonds to be valued; inputting the sparse characteristics into a bond valuation model, and obtaining a future transaction price valuation of the bond to be valued based on an output value of the bond valuation model; and displaying the price valuation of the future deal through a bond price display page. In the embodiment, the evaluation model of the bond is trained by using the historical transaction price of the bond to be evaluated and the historical transaction price of the bond of the same type, and the evaluation of the bond is performed by using the evaluation model of the bond based on the historical price of the bond to be evaluated and the real-time transaction price of the bond of the same type of the bond to be evaluated, so that the real-time evaluation can be realized, and the evaluation result is more accurate.
The foregoing description is only an overview of the technical solutions of the present application, and the following detailed description of the present application is given to enable the technical means of the present application to be more clearly understood and to enable the above and other objects, features, and advantages of the present application to be more clearly understood.
Drawings
In the drawings, like reference characters designate like or similar parts or elements throughout the several views unless otherwise specified. The figures are not necessarily to scale. It is appreciated that these drawings depict only some embodiments in accordance with the application and are not to be considered limiting of its scope.
Fig. 1 is a schematic view of a scenario of a bond valuation method according to an embodiment of the present application;
fig. 2 is a flowchart of a bond valuation method according to an embodiment of the present application;
FIG. 3 is a flowchart of a method for training a valuation model of bonds according to an embodiment of the present application;
fig. 4 is a schematic diagram of a bond valuation method according to an embodiment of the present application;
fig. 5 is a block diagram illustrating a structure of a bond valuation apparatus according to an embodiment of the present application;
fig. 6 is a block diagram illustrating a structure of a training apparatus for a bond valuation model according to an embodiment of the present application; and
FIG. 7 is a block diagram of an electronic device used to implement embodiments of the present application.
Detailed Description
In the following, only certain exemplary embodiments are briefly described. As those skilled in the art will recognize, the described embodiments may be modified in various different ways, without departing from the spirit or scope of the present application. Accordingly, the drawings and description are to be regarded as illustrative in nature, and not as restrictive.
In order to facilitate understanding of the technical solutions of the embodiments of the present application, the following description is made of related art of the embodiments of the present application. The following related technologies may be arbitrarily combined with the technical solutions of the embodiments of the present application as alternatives, which all belong to the scope of protection of the embodiments of the present application.
Fig. 1 is a schematic diagram of an exemplary application scenario for implementing the method of the embodiment of the present application. The cloud computing platform provides cloud services for tenants, the cloud services can be deployed in tenant equipment in a virtual machine mode, and functions of the cloud services can include evaluation of bonds. The cloud computing server obtains a training sample set, the training sample set comprises training samples and sample labels, the training samples comprise sparse features obtained based on historical transaction prices of bonds to be evaluated and historical transaction prices of bonds of the same type, and the sample labels comprise differences between the historical transaction prices of the bonds to be evaluated and the historical transaction prices of the bonds of the same type. And training the neural network model by utilizing the training sample set to obtain a bond valuation model of the bond to be valued. Wherein the same type includes at least one of: the bond ratings are the same, the bond expiry dates are the same, the industries to which the bond issuers belong are the same, and the districts to which the bond issuers belong are the same.
When the evaluation is carried out on the bond, the sparse characteristic of the bond to be evaluated is determined based on the historical price of the bond to be evaluated and the real-time transaction price of the bond of the same type of the bond to be evaluated; inputting the sparse characteristics into a bond valuation model, obtaining a future transaction price valuation of the bond to be valued based on an output value of the bond valuation model, sending the future transaction price valuation to tenant equipment, and displaying the tenant equipment through a bond price display page of a cloud service client side for a bond trader to quote or providing reference for a user who intentionally purchases the bond.
In the related art, bonds whose deal data is sparse, for example, credit bonds. Rating credit and debt based on inherent characteristics of the credit and debt, collecting bargaining prices of the credit and debt of a key term, drawing a yield curve for the same rated credit and debt based on an Hermite difference method, and estimating the credit and debt based on the yield curve. The estimation mode can only estimate in units of days, cannot estimate in real time, needs experienced manual intervention for rating credit debt and is low in efficiency.
In the embodiment, the historical transaction price of the bond of the same type is adopted as the training sample of the bond valuation model, so that the problem of sparse historical transaction data of the bond to be valued can be solved; and constructing sparse characteristics for valuation based on the historical price of the bond to be valued and the real-time bargain price of the same type of bond of the bond to be valued, so that real-time valuation can be realized, and the valuation result is more accurate.
An embodiment of the present application provides a bond valuation method, as shown in fig. 2, which is a flowchart of the bond valuation method according to an embodiment of the present application, and an execution subject of the method may be a computing device, including: servers, user devices, etc. The method comprises the following steps:
step S201, determining sparse characteristics of the bonds to be evaluated based on the historical prices of the bonds to be evaluated and the real-time transaction prices of the bonds of the same type of the bonds to be evaluated.
The bonds to be evaluated may include bonds with sparse historical transaction prices, such as credit bonds, bonds issued by other than the government and promised to pay cash flow with certain interest. The bond to be evaluated may be other bonds, and the application is not limited thereto.
The historical price of the bond to be evaluated can be a historical evaluation price or a historical transaction price. For example, the price on the day of the credit debt is estimated, but the credit debt was not committed, i.e., no committed price, the historical price may be the valuation price of the previous day.
The same type of bond in the same type of bond includes at least one of: the bond ratings are the same, the bond expiration dates are the same, the industries to which the bond issuers belong are the same, and the regions to which the bond issuers belong are the same. The bond rating may include a bond credit rating, i.e., a credit rating performed on the basis of a valuable bond issued by an enterprise or an economic entity. The credit rating of the bond is mostly the credit rating of the bond of the enterprise, and is used for evaluating the reliability of paying the money according to the time and indicating the credit rating of a specific bond issued by the enterprise with the independent legal qualification. Such a credit rating may provide information services for negotiable transfer campaigns where investors purchase bonds and bond in stock market. The bond expiration period may be a period during which the bond can be redeemed.
The sparse feature refers to data represented by a group of multidimensional vectors, a part of components of the vectors are all 0, and the number of nonzero components is smaller than the dimensionality of the multidimensional vectors.
And determining sparse characteristics based on the historical prices of the bonds to be evaluated and the real-time bargaining prices of the bonds of the same type of the bonds to be evaluated, wherein some components are 0, and the components are defaulted to be 0 because some bonds of the same type are not bargained and the real-time bargaining prices do not exist.
The real-time transaction price of the bond of the same type can be the latest transaction price, for example, the bond to be evaluated is evaluated in real time at 3 pm on the same day, if no transaction is made at the day before the bond to be evaluated, the sparse feature of the bond to be evaluated is constructed by using the evaluation of the day before and the transaction price of the bond of the same type which has been committed at 3 pm on the same day.
And S202, inputting the sparse characteristics into a bond valuation model, and obtaining a future transaction price valuation of the bond to be valued based on an output value of the bond valuation model.
The bond valuation model is obtained by training through the historical deal price of the bond to be valued and the historical deal price of the bond of the same type. The estimation prediction model may be trained based on a neural network model, for example, the neural network model may be a piecewise linear neural network model PLNN.
And step S203, displaying the future transaction price estimated value through a bond price display page.
The present embodiment is described with a server as an execution subject. The server sends the future transaction price valuation to the terminal equipment of the tenant, so that the user can check the future transaction price valuation through the bond price display page of the cloud service client.
The bond valuation method provided by the embodiment of the application determines the sparse characteristic of the bond to be valued based on the historical price of the bond to be valued and the real-time transaction price of the bond of the same type of the bond to be valued; inputting the sparse characteristics into a bond valuation model, and obtaining a future transaction price valuation of the bond to be valued based on an output value of the bond valuation model; and displaying the price estimated value of the future deal through a bond price display page. In the embodiment, the evaluation model of the bond is trained by using the historical transaction price of the bond to be evaluated and the historical transaction price of the bond of the same type, and the evaluation of the bond is performed by using the evaluation model of the bond based on the historical price of the bond to be evaluated and the real-time transaction price of the bond of the same type of the bond to be evaluated, so that the real-time evaluation can be realized, and the evaluation result is more accurate.
The specific implementation manner of constructing the sparse feature in step S201 is shown in the following embodiments:
in one possible implementation manner, the determining the sparse feature of the bond to be evaluated based on the historical price of the bond to be evaluated and the real-time transaction price of the bond of the same type of the bond to be evaluated comprises the following steps: and (3) making a difference between the historical price of the bond to be evaluated and the real-time transaction prices of a plurality of bonds of the same type of the bond to be evaluated, and determining the sparse characteristic of the bond to be evaluated based on a plurality of difference values.
In practical applications, the bonds to be evaluated of the same type can be multiple, and according to the inherent characteristics of the bonds to be evaluated: the rating of the bond, the expiration date of the bond, the industry to which the issuer of the bond belongs, the territory to which the issuer of the bond belongs, etc., a plurality of bonds of the same type can be determined. For example, there may be a plurality of bonds with the same rating, a plurality of bonds with the same industry to which the bond issuer belongs, a plurality of bonds with the same evaluation and a plurality of industries to which the issuer belongs, and the historical price of the bond to be evaluated is differentiated from the real-time bargain price of the bond of the same type to obtain a plurality of difference values as elements in the feature matrix, thereby obtaining the sparse feature of the bond to be evaluated.
Optionally, a mean value of the plurality of difference values may be calculated as an element in the feature matrix, for example, a plurality of difference values of the historical price of the bond to be evaluated and the real-time transaction price of the bond with the same credit rating are calculated, a mean value of the plurality of difference values is calculated, and the feature matrix is constructed.
In the embodiment, the sparse feature is constructed for valuation based on the historical price of the bond to be valued and the real-time transaction price of the same type of bond of the bond to be valued, so that real-time valuation can be realized, and the valuation result is more accurate.
In one possible implementation, obtaining an estimate of future deal price of the bond to be evaluated based on the output value of the bond evaluation model comprises: taking the output value of the bond valuation model as the future transaction price valuation of the bond to be valued; or summing the output value of the bond valuation model and the historical price of the bond to be valued to obtain the future transaction price valuation of the bond to be valued.
The bond valuation model can directly output future transaction price valuation of the bond to be valued according to the input sparse characteristics; or the bond valuation model directly predicts the difference between the historical price of the bond to be valued and the real-time price of the bond of the same type according to the input sparse features, and sums the predicted difference and the historical price of the bond to be valued to obtain the future deal price valuation. It is understood that the bond valuation model can obtain different output values according to the structure of the model.
Illustratively, a summation calculation module is added in the bond valuation model and used for calculating the sum of the predicted difference and the historical price of the bond to be valued, so that the bond valuation model can directly output the future deal price valuation of the bond to be valued according to the input sparse features.
Illustratively, a summation calculation module is added outside the bond valuation model, the output value of the bond valuation model is the difference value between the predicted historical price of the bond to be valued and the real-time price of the bond of the same type, and the summation calculation module is used for calculating the sum of the predicted difference value and the historical price of the bond to be valued to obtain the future deal price valuation.
An embodiment of the present application provides a method for training a bond valuation model, and as shown in fig. 3, a flowchart of the method for training a bond valuation model according to an embodiment of the present application is shown, where an execution subject of the method may be a computing device, and the method includes: servers, user terminals, etc. The method comprises the following steps:
step S301, a training sample set is obtained, the training sample set comprises training samples and sample labels, the training samples comprise sparse features obtained based on historical transaction prices of the bonds to be evaluated and historical transaction prices of the bonds of the same type, and the sample labels comprise differences between the historical transaction prices of the bonds to be evaluated and the historical transaction prices of the bonds of the same type.
The historical transaction price of the bond to be evaluated is the historical actual transaction price of the bond to be evaluated, the historical actual transaction price of the bond to be evaluated and the historical transaction prices of bonds of the same type are utilized, a sparse matrix is constructed to serve as a training sample to train a neural network model, and the real-time future transaction price evaluation of the bond to be evaluated can be obtained. The historical transaction price of the same type of bond is used as a training sample, so that the problem of sparse transaction data of the bond to be evaluated can be solved, and the result obtained by training is more accurate.
And step S302, training the neural network model by using the training sample set to obtain a bond evaluation model of the bond to be evaluated.
The estimation prediction model may be trained based on a neural network model, for example, the neural network model may be a piecewise linear neural network model PLNN.
The method for training the bond valuation model obtains a training sample set, wherein the training sample set comprises training samples and sample labels, the training samples comprise sparse features obtained based on historical transaction prices of bonds to be valuation and historical transaction prices of bonds of the same type, and the sample labels comprise differences between the historical transaction prices of the bonds to be valuation and the historical transaction prices of the bonds of the same type; and training the neural network model by using the training sample set to obtain a bond evaluation model of the bond to be evaluated. In the embodiment, the evaluation model of the bond is trained by using the historical transaction price of the bond to be evaluated and the historical transaction price of the bond of the same type, and the evaluation of the bond is performed by using the evaluation model of the bond based on the historical price of the bond to be evaluated and the real-time transaction price of the bond of the same type of the bond to be evaluated, so that the real-time evaluation can be realized, and the evaluation result is more accurate.
In one possible implementation, the method further includes: and adjusting the activation interval of the activation function of the bond valuation model, and training the bond valuation model based on the adjusted activation function.
In the model training process, the activation function of the bond valuation model can be adjusted, so that the loss of the component 0 in the sparse feature is not transmitted back, and the prediction accuracy of the bond valuation model is improved.
In one example, the activation interval of the activation function is adjusted according to the following formula:
Figure BDA0003865241980000061
Figure BDA0003865241980000062
wherein, f 1 (x) Indicating the activation function before adjustment, f 2 (x) Indicating the activation function after adjustment. x represents a component in the sparse feature. Epsilon is an empirical value and can be matched according to specific needsAnd (4) placing. Before the function is adjusted, as shown in formula (1), the neural network model automatically learns the piecewise function, and is activated more than 0 and not activated less than 0. After the activation function is adjusted, as shown in formula (2), the segmentation point is 0, and is not activated at the 0 point, and is activated at the non-0 point.
In one possible implementation, the sparse feature is obtained by: and respectively differentiating the plurality of historical trading prices of the bond to be evaluated with the plurality of historical trading prices of the bond of the same type, and determining the sparse characteristic based on the plurality of differential values.
In practical application, the bond to be evaluated can be a plurality of bonds of the same type, and according to the inherent characteristics of the bond to be evaluated: the rating of the bond, the expiration date of the bond, the industry to which the bond issuer belongs, the region to which the bond issuer belongs, etc., a plurality of bonds of the same type can be determined. For example, there may be a plurality of bonds with the same rating, a plurality of bonds with the same industry to which the bond issuer belongs, a plurality of bonds with the same evaluation and a plurality of industries to which the issuer belongs, and the historical transaction prices of the bonds to be evaluated and the historical transaction prices of the bonds of the same type are differentiated to obtain a plurality of difference values, which are used as elements in the feature matrix to obtain the training samples.
In the embodiment, sparse features are constructed to serve as training samples based on the historical deal prices of the bonds to be evaluated and the historical deal prices of the bonds of the same type of the bonds to be evaluated, the historical deal prices of the bonds of the same type are used as the training samples, the problem of sparse deal data of the bonds to be evaluated can be solved, and the result obtained by training is more accurate.
In order to more clearly introduce the technical idea of the technical solution of the present application, the method provided by the present application is described below by a specific embodiment. Fig. 4 is a schematic diagram illustrating a bond valuation method according to an embodiment of the present application. In this embodiment, the bond to be evaluated is explained by taking a credit bond as an example. As shown in fig. 4, first, the inherent features of the credit debt are acquired, including: and the real-time transaction information of the bonds of the same type is acquired by inquiring the bonds of the same type according to the inherent characteristics, such as bond rating, bond expiration date, industry to which a bond issuer belongs, region to which the bond issuer belongs, and the like. And taking the previous day valuation of the credit bonds as discrete characteristics, and carrying out characteristic construction according to the discrete characteristics and the real-time transaction information of bonds of the same type to obtain the real-time characteristics of similar bonds, and representing the real-time fluctuation condition of the credit bonds of the same type on the same day. Secondly, inputting the real-time characteristics of the similar bonds into a bond valuation model obtained in advance based on PLNN training, and performing prediction valuation by using the bond valuation model to obtain a future transaction price valuation of the credit bonds. The future transaction price valuation is displayed through a bond price display page for the bond trader to quote or provide reference for the user who is interested in purchasing bonds.
Corresponding to the application scenario and method of the method provided by the embodiment of the application, the embodiment of the application further provides a bond valuation device. Fig. 5 is a block diagram illustrating a structure of a bond valuation apparatus according to an embodiment of the present application, which may include:
the characteristic determining module 501 is configured to determine sparse characteristics of the bonds to be evaluated based on historical prices of the bonds to be evaluated and real-time transaction prices of bonds of the same type of the bonds to be evaluated;
the price determination module 502 is used for inputting the sparse characteristics into the bond valuation model, and obtaining a future transaction price valuation of the bond to be valued based on the output value of the bond valuation model;
the price display module 503 is configured to display the future transaction price valuation through a bond price display page;
the bond valuation model is obtained by training by utilizing the historical trading price of the bond to be valued and the historical trading price of the bond of the same type.
The bond valuation device provided by the embodiment of the application determines the sparse characteristic of the bond to be valued based on the historical price of the bond to be valued and the real-time transaction price of the bond of the same type of the bond to be valued; inputting the sparse characteristics into a bond valuation model, and obtaining a future transaction price valuation of the bond to be valued based on an output value of the bond valuation model; and displaying the price valuation of the future deal through a bond price display page. In the embodiment, the bond valuation model is trained by using the historical deal price of the bond to be valuated and the historical deal price of the bond of the same type, and the bond valuation model is used for carrying out bond valuation based on the historical price of the bond to be valuated and the real-time deal price of the bond of the same type of the bond to be valuated, so that real-time valuation can be realized, and valuation results are more accurate.
In a possible implementation manner, the feature determining module 501 is specifically configured to:
and (3) making a difference between the historical price of the bond to be evaluated and the real-time transaction prices of a plurality of bonds of the same type of the bond to be evaluated, and determining the sparse characteristic of the bond to be evaluated based on a plurality of difference values.
In one possible implementation, the price determination module 502, when obtaining an estimate of a future deal price of a bond to be evaluated based on the output values of the bond evaluation model, is configured to:
taking the output value of the bond valuation model as a future transaction price valuation of the bond to be valued; or
And summing the output value of the bond valuation model and the historical price of the bond to be valued to obtain the future transaction price valuation of the bond to be valued.
In one possible implementation, the same type includes at least one of:
the bond ratings are the same, the bond expiration dates are the same, the industries to which the bond issuers belong are the same, and the regions to which the bond issuers belong are the same.
The functions of each module in each device in the embodiment of the present application can be referred to the corresponding description in the above method, and have corresponding beneficial effects, which are not described herein again.
Corresponding to the application scenario and the method of the method provided by the embodiment of the application, the embodiment of the application further provides a device for training the bond valuation model. Fig. 6 is a block diagram illustrating a structure of a bond valuation model training device according to an embodiment of the present application, where the device may include:
the sample acquisition module 601 is configured to acquire a training sample set, where the training sample set includes training samples and sample labels, the training samples include sparse features obtained based on historical transaction prices of bonds to be evaluated and historical transaction prices of bonds of the same type, and the sample labels include differences between the historical transaction prices of the bonds to be evaluated and the historical transaction prices of the bonds of the same type;
and the model training module 602 is configured to train the neural network model by using the training sample set to obtain a bond evaluation model of the bond to be evaluated.
The method for training the bond valuation model, provided by the embodiment of the application, comprises the steps of obtaining a training sample set, wherein the training sample set comprises training samples and sample labels, the training samples comprise sparse features obtained based on historical transaction prices of bonds to be valued and historical transaction prices of bonds of the same type, and the sample labels comprise the difference between the historical transaction prices of the bonds to be valued and the historical transaction prices of the bonds of the same type; and training the neural network model by utilizing the training sample set to obtain a bond valuation model of the bond to be valued. In the embodiment, the evaluation model of the bond is trained by using the historical transaction price of the bond to be evaluated and the historical transaction price of the bond of the same type, and the evaluation of the bond is performed by using the evaluation model of the bond based on the historical price of the bond to be evaluated and the real-time transaction price of the bond of the same type of the bond to be evaluated, so that the real-time evaluation can be realized, and the evaluation result is more accurate.
In one possible implementation manner, the apparatus further includes a model adjusting module configured to:
and adjusting the activation interval of the activation function of the bond valuation model, and training the bond valuation model based on the adjusted activation function.
In one possible implementation, the sparse feature is obtained by:
and respectively differentiating the plurality of historical transaction prices of the bond to be evaluated with the historical transaction prices of the plurality of bonds of the same type, and determining the sparse characteristic based on the plurality of differential values.
The functions of each module in each device in the embodiment of the present application can be referred to the corresponding description in the above method, and have corresponding beneficial effects, which are not described herein again.
FIG. 7 is a block diagram of an electronic device used to implement embodiments of the present application. As shown in fig. 7, the electronic apparatus includes: a memory 710 and a processor 720, the memory 710 having stored therein computer programs that are executable on the processor 720. The processor 720, when executing the computer program, implements the methods in the embodiments described above. The number of the memory 710 and the processor 720 may be one or more.
The electronic device further includes:
and a communication interface 730, configured to communicate with an external device, and perform data interactive transmission.
If the memory 710, the processor 720 and the communication interface 730 are implemented independently, the memory 710, the processor 720 and the communication interface 730 may be connected to each other through a bus and perform communication with each other. The bus may be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, an Extended ISA (EISA) bus, or the like. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one thick line is shown in FIG. 7, but this is not intended to represent only one bus or type of bus.
Optionally, in an implementation, if the memory 710, the processor 720 and the communication interface 730 are integrated on a chip, the memory 710, the processor 720 and the communication interface 730 may complete communication with each other through an internal interface.
Embodiments of the present application provide a computer-readable storage medium, which stores a computer program, and when the program is executed by a processor, the computer program implements the method provided in the embodiments of the present application.
The embodiment of the present application further provides a chip, where the chip includes a processor, and is configured to call and run an instruction stored in a memory from the memory, so that a communication device in which the chip is installed executes the method provided in the embodiment of the present application.
An embodiment of the present application further provides a chip, including: the system comprises an input interface, an output interface, a processor and a memory, wherein the input interface, the output interface, the processor and the memory are connected through an internal connection path, the processor is used for executing codes in the memory, and when the codes are executed, the processor is used for executing the method provided by the embodiment of the application.
It should be understood that the Processor may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic, discrete hardware components, etc. A general purpose processor may be a microprocessor or any conventional processor or the like. It is noted that the processor may be a processor supporting an Advanced reduced instruction set machine (ARM) architecture.
Further, optionally, the memory may include a read-only memory and a random access memory. The memory may be either volatile memory or nonvolatile memory, or may include both volatile and nonvolatile memory. The non-volatile Memory may include a Read-Only Memory (ROM), a Programmable ROM (PROM), an Erasable PROM (EPROM), an Electrically Erasable PROM (EEPROM), or a flash Memory. Volatile Memory can include Random Access Memory (RAM), which acts as external cache Memory. By way of example, and not limitation, many forms of RAM may be used. For example, static Random Access Memory (Static RAM, SRAM), dynamic Random Access Memory (DRAM), synchronous Dynamic Random Access Memory (SDRAM), double Data Rate Synchronous Dynamic Random Access Memory (DDR SDRAM), enhanced SDRAM (ESDRAM), SLDRAM (SLDRAM), and Direct Rambus RAM (DR RAM).
In the above embodiments, the implementation may be wholly or partially realized by software, hardware, firmware, or any combination thereof. When implemented in software, it may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. The procedures or functions according to the present application are all or partially generated when the computer program instructions are loaded and executed on a computer. The computer may be a general purpose computer, a special purpose computer, a network of computers, or other programmable device. The computer instructions may be stored in a computer readable storage medium or transmitted from one computer readable storage medium to another computer readable storage medium.
In the description herein, references to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the application. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Moreover, various embodiments or examples and features of various embodiments or examples described in this specification can be combined and combined by one skilled in the art without being mutually inconsistent.
Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or to implicitly indicate the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of the present application, "a plurality" means two or more unless specifically limited otherwise.
Any process or method described in a flowchart or otherwise herein may be understood as representing a module, segment, or portion of code, which includes one or more executable instructions for implementing specific logical functions or steps of the process. And the scope of the preferred embodiments of the present application includes other implementations in which functions may be performed out of the order shown or discussed, including substantially concurrently or in reverse order, depending on the functionality involved.
The logic and/or steps described in the flowcharts or otherwise described herein, such as an ordered listing of executable instructions that can be considered to implement logical functions, can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions.
It should be understood that portions of the present application may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the various steps or methods may be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. All or a portion of the steps of the method of the above embodiments may be performed by associated hardware that is instructed by a program, which may be stored in a computer-readable storage medium, that when executed, includes one or a combination of the steps of the method embodiments.
In addition, functional units in the embodiments of the present application may be integrated into one processing module, or each unit may exist alone physically, or two or more units are integrated into one module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode. The integrated module may also be stored in a computer-readable storage medium if it is implemented in the form of a software functional module and sold or used as a separate product. The storage medium may be a read-only memory, a magnetic or optical disk, or the like.
The above description is only an exemplary embodiment of the present application, but the scope of the present application is not limited thereto, and any person skilled in the art can easily conceive various changes or substitutions within the technical scope described in the present application, and these should be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims (10)

1. A method for bond valuation, said method comprising:
determining the sparse characteristic of the bond to be evaluated based on the historical price of the bond to be evaluated and the real-time transaction price of the bond of the same type of the bond to be evaluated;
inputting the sparse features into a bond valuation model, and obtaining a future transaction price valuation of the bond to be valued based on an output value of the bond valuation model;
displaying the future transaction price valuation through a bond price display page;
the bond valuation model is obtained by training through the historical deal price of the bond to be valued and the historical deal price of the bond of the same type.
2. The method of claim 1, wherein determining sparse features of the bonds to be evaluated based on historical prices of the bonds to be evaluated and real-time closing prices of bonds of the same type of the bonds to be evaluated comprises:
and making a difference between the historical price of the bond to be evaluated and the real-time transaction price of a plurality of bonds of the same type of the bond to be evaluated, and determining the sparse characteristic of the bond to be evaluated based on a plurality of difference values.
3. The method of claim 1, wherein obtaining an estimate of future closing prices of the bonds to be valued based on the output values of the bond valuation model comprises:
taking the output value of the bond valuation model as the future transaction price valuation of the bond to be valued; or
And summing the output value of the bond valuation model and the historical price of the bond to be valued to obtain the future transaction price valuation of the bond to be valued.
4. The method according to any of claims 1-3, wherein the same type comprises at least one of:
the bond ratings are the same, the bond expiration dates are the same, the industries to which the bond issuers belong are the same, and the regions to which the bond issuers belong are the same.
5. A method for training a bond valuation model, the method comprising:
acquiring a training sample set, wherein the training sample set comprises training samples and sample labels, the training samples comprise sparse features obtained based on historical transaction prices of bonds to be evaluated and historical transaction prices of bonds of the same type, and the sample labels comprise differences between the historical transaction prices of the bonds to be evaluated and the historical transaction prices of the bonds of the same type;
and training a neural network model by using the training sample set to obtain a bond valuation model of the bond to be valued.
6. The method of claim 5, further comprising:
and adjusting the activation interval of the activation function of the bond valuation model, and training the bond valuation model based on the adjusted activation function.
7. The method according to claim 5 or 6, wherein the sparse feature is obtained by:
and respectively differentiating the plurality of historical trading prices of the bond to be evaluated with the plurality of historical trading prices of the bond of the same type, and determining the sparse feature based on the plurality of differential values.
8. A bond valuation apparatus, said apparatus comprising:
the characteristic determination module is used for determining the sparse characteristic of the bond to be evaluated based on the historical price of the bond to be evaluated and the real-time transaction price of the bond of the same type of the bond to be evaluated;
the price determination module is used for inputting the sparse characteristics into a bond valuation model and obtaining a future transaction price valuation of the bond to be valued based on an output value of the bond valuation model;
the price display module is used for displaying the future transaction price valuation through a bond price display page;
the bond valuation model is obtained by training through the historical deal price of the bond to be valued and the historical deal price of the bond of the same type.
9. A bond valuation model training apparatus, said apparatus comprising:
the system comprises a sample acquisition module, a comparison module and a comparison module, wherein the sample acquisition module is used for acquiring a training sample set, the training sample set comprises training samples and sample labels, the training samples comprise sparse features obtained based on historical transaction prices of bonds to be evaluated and historical transaction prices of bonds of the same type, and the sample labels comprise the difference between the historical transaction prices of the bonds to be evaluated and the historical transaction prices of the bonds of the same type;
and the model training module is used for training the neural network model by utilizing the training sample set to obtain a bond evaluation model of the bond to be evaluated.
10. An electronic device comprising a memory, a processor and a computer program stored on the memory, the processor implementing the method of any one of claims 1-7 when executing the computer program.
CN202211177561.XA 2022-09-26 2022-09-26 Bond valuation method, bond valuation model training method and device Pending CN115482041A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202211177561.XA CN115482041A (en) 2022-09-26 2022-09-26 Bond valuation method, bond valuation model training method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202211177561.XA CN115482041A (en) 2022-09-26 2022-09-26 Bond valuation method, bond valuation model training method and device

Publications (1)

Publication Number Publication Date
CN115482041A true CN115482041A (en) 2022-12-16

Family

ID=84394698

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202211177561.XA Pending CN115482041A (en) 2022-09-26 2022-09-26 Bond valuation method, bond valuation model training method and device

Country Status (1)

Country Link
CN (1) CN115482041A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116630051A (en) * 2023-07-24 2023-08-22 中债金融估值中心有限公司 Real-time calculation system, method and equipment for yield curve and bond estimation

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116630051A (en) * 2023-07-24 2023-08-22 中债金融估值中心有限公司 Real-time calculation system, method and equipment for yield curve and bond estimation

Similar Documents

Publication Publication Date Title
JP4464281B2 (en) Systems and methods for buying and selling higher or lower than the market price
Abid et al. The determinants of credit default swap rates: an explanatory study
Brown et al. The interaction between order imbalance and stock price
US20080243717A1 (en) Method and system for managing a mortgage-backed securities index
US7885892B1 (en) Method and system for assessing repurchase risk
US20140316970A1 (en) Generating income from unused credit
KR20180118922A (en) Lending Meditation Platform System and Credit Estimating Apparatus and Method thereof
Garvey et al. The realised–implied volatility relationship: Recent empirical evidence from FTSE‐100 Stocks
US20220138859A1 (en) Method of Executing Orders Using an Electronic Forum
KR101444249B1 (en) Method, system and non-transitory computer-readable recording medium for providing information on securities lending and borrowing transaction, short selling or equity swap transaction
CN115482041A (en) Bond valuation method, bond valuation model training method and device
Agliardi A comprehensive structural model for defaultable fixed-income bonds
US20040215539A1 (en) Foreign exchange covered warrant system and structure
Ingermann et al. The recovery rate for retail and commercial customers in Germany: a look at collateral and its adjusted market values
Narayanan et al. Credit insurance, distress resolution costs, and bond spreads
US8407123B2 (en) Methods and systems for providing a yield curve index
WO2011059483A1 (en) Price sensitivity scores
US20080154786A1 (en) Single party platform for sale and settlement of OTC derivatives
EP3783560A1 (en) Automated objective generation of data for, and post validation of, estimation of term sofr benchmarks
CN111967987B (en) Transaction data processing method and transaction data processing system
US20160307245A1 (en) Computer implemented systems and methods for asset transfer
Chen et al. Commercial Mortgage‐Backed Security Pricing with Real Estate Liquidity Risk
CN110909294A (en) Data processing method and device
Huang The pricing of conditional performance guarantees with risky collateral
US20220076330A1 (en) Method, apparatus, and computer readable medium for generating a real-time risk score associated with financing of an invoice based on real-time transaction data

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination