Bond yield prediction method and device, equipment and storage medium

Abstract

The invention relates to the technical field of scene prediction, in particular to a bond yield prediction method and device, equipment and a storage medium, and the method comprises the steps: determining an ultimate long-term interest rate based on historical bond data and a preset Kalman filter; processing the historical bond data based on the ultimate long-term interest rate and a Smith-Wilson model to determine autoregression parameters; calculating bond price simulation statistics based on a predicted bond yield obtained by processing the historical bond data by the Kalman filter; and calculating a bond yield based on the autoregression parameter and bond price simulation statistics. The method is convenient for improving the prediction accuracy of the bond yield.

Description

Technical Field

[0001] The present application relates to the technical field of scenario prediction, and in particular to a bond yield prediction method, device, equipment and storage medium. Background Art

[0002] As bank interest rates continue to fall, the problem of insurance companies' minimum interest rate guarantee can easily turn from a problem of capital shortage to a problem of capital maturity, which can easily lead to insurance companies facing solvency issues. To deal with this problem, it is necessary to ensure that the valuation of insurance companies' assets and liabilities is as consistent as possible with the market, especially in terms of bond yield forecasts.

[0003] At present, in terms of predicting bond yields, an affine term structure model explored by relevant researchers has received great attention. This affine term structure model is used to predict bond yields.

[0004] However, when using this affine term structure model to predict bond yields, it is difficult to achieve accurate fitting of sample data and accurate estimation of model parameters, resulting in poor prediction accuracy of bond yields using this affine term structure model. Summary of the invention

[0005] In order to improve the prediction accuracy of bond yields, the present application provides a bond yield prediction method, device, equipment and storage medium.

[0006] In a first aspect, the present application provides a bond yield prediction method, comprising:

[0007] Determine the ultimate forward rate based on historical bond data and a preset Kalman filter;

[0008] Determine autoregressive parameters by processing the historical bond data based on the terminal forward rate and the Smith-Wilson model;

[0009] Calculating bond price simulation statistics based on the predicted bond yield obtained by processing the historical bond data with the Kalman filter;

[0010] The bond yield is calculated based on the autoregressive parameters and the bond price simulation statistics.

[0011] In a second aspect, the present application provides a bond yield prediction device, comprising:

[0012] An interest rate determination module, used to determine the ultimate forward interest rate based on historical bond data and a preset Kalman filter;

[0013] A parameter determination module, used for determining autoregressive parameters by processing the historical bond data based on the ultimate forward rate and the Smith-Wilson model;

[0014] A statistics calculation module, used for calculating bond price simulation statistics based on the predicted bond yield obtained by processing the historical bond data by the Kalman filter;

[0015] The yield prediction module is used to calculate the bond yield based on the autoregressive parameter and the bond price simulation statistics.

[0016] In a third aspect, the present application provides a computer device, the computer device comprising a memory and a processor, the memory storing a computer program, and the processor implementing the steps in the above method when executing the computer program.

[0017] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which implements the steps in the above method when executed by a processor.

[0018] In a fifth aspect, the present application further provides a computer program product, wherein the computer program product comprises a computer program, and when the computer program is executed by a processor, the steps in any of the above method embodiments are implemented.

[0019] The above-mentioned bond yield prediction method, device, equipment and storage medium determine the ultimate forward interest rate based on historical bond data and a preset Kalman filter; determine the autoregressive parameter based on the ultimate forward interest rate and the Smith-Wilson model to process the historical bond data; calculate the bond price simulation statistic based on the predicted bond yield obtained by processing the historical bond data with the Kalman filter; and calculate the bond yield based on the autoregressive parameter and the bond price simulation statistic. Through the above implementation, not only can the accurate fitting of historical bond data be achieved, but also the accurate estimation of parameters can be achieved, which is convenient for improving the prediction accuracy of bond yields.

[0020] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0022] Figure 1 A flow chart of a bond yield prediction method provided in an embodiment of the present application;

[0023] Figure 2 This is a schematic diagram of the structure of a bond yield prediction device provided in an embodiment of the present application;

[0024] Figure 3 A schematic diagram of the structure of a computer device provided in an embodiment of the present application;

[0025] Figure 4 This is a diagram of the internal structure of a computer-readable storage medium provided in an embodiment of the present application. DETAILED DESCRIPTION

[0026] In order to make the purpose, technical solution and advantages of the present disclosure more clear, the present disclosure is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present disclosure and are not used to limit the present disclosure.

[0027] It should be noted that the terms "first", "second", etc. in the specification and claims of this article and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of this article described here can be implemented in an order other than those illustrated or described here. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, device, product or equipment that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or equipment.

[0028] In this article, the term "and / or" is only a description of the association relationship between related objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the related objects before and after are in an "or" relationship.

[0029] Embodiment 1

[0030] Figure 1 A flow chart of a bond yield prediction method provided in Example 1 of the present application, referring to Figure 1 The method may be performed by a device for performing the method, and the device may be implemented by software and / or hardware. The method includes:

[0031] S110. Determine the ultimate forward interest rate based on historical bond data and a preset Kalman filter.

[0032] In this embodiment, the bonds are specifically treasury bonds as an example, and in other embodiments, there is no specific limitation; in order to realize the prediction of bond yields, it is necessary to first determine the ultimate forward interest rate, which is the long-term interest rate related to the bonds; in this embodiment, the dynamic Nelson-Siegel model (DNS) can be used for the ultimate forward interest rate, and the expression of the dynamic Nelson-Siegel model is as follows:

[0033]

[0034] Among them, τ is the maturity date of the bond, y t is the bond yield corresponding to τ, L t is the ultimate forward interest rate. When τ approaches infinity, y t (τ) approaches L t ; S t To affect the slope of the bond yield curve, C t It is the influencing factor of the curvature of the bond yield curve, which is used to determine the curvature of the yield curve; λ is the attenuation factor.

[0035] In addition, to simplify the expression of the dynamic Nelson-Siegel model, the expression can be rewritten into a vector form, that is: t =Γ(λ)β t +ε t ; ε t Obeys normal independent distribution; y t =[y t (τ 1 ),…,y t (τ N )]',τ N Indicates the nth expiration date, where y t is the bond yield vector; Γ(λ) is the factor loading matrix, β t is the latent vector, ε t is the perturbation vector, and ε t It obeys normal independent distribution; and the specific expression of the factor loading matrix is ​​as follows:

[0036]

[0037] Among them, i represents the number of columns in the factor loading matrix, and j represents the number of rows in the factor loading matrix.

[0038] It should be noted that the above formula "y t =Γ(λ)β t +ε t", which corresponds to the observation process equation of the Kalman filter; the Kalman filter also includes a prediction process equation. This embodiment uses a first-order vector autoregression process (VAR (1) model) to construct the prediction process equation. The prediction process equation is:

[0039] β t =(I-Φ)μ+Φβ t-1 +η t ;

[0040] Among them, I is the identity matrix, Φ is the coefficient matrix, μ is the mean vector, η t is the residual error β of the prediction process t The prediction process and observation value y can be obtained through the Kalman filter 1 ,…,y t Estimation is performed and β t The lowest mean square linear estimator is denoted as b t∣s , and also β t The mean square error matrix is ​​recorded as B t∣s , where s = t-1, t; during the update process of the Kalman filter, b t∣s With B t∣s The calculation formula is as follows:

[0041]

[0042] Among them, K t is the Kalman gain, v t is the prediction error vector, F t is the forecast error variance matrix; and:

[0043]

[0044] Among them, Σ ε is the variance of the observed process residual;

[0045]

[0046] Among them, Σ η is the variance of the residual of the forecast process; historical bond data includes b t∣s With B t∣s Initialization data, b t∣s The initialization data is b 1∣0 =μ, B t∣s The initialization data is B 1∣0 =Σ β ,Σ β β t The variance of .

[0047] Specifically, the Kalman filter processes historical bond data through its prediction process and observation process to determine the ultimate forward interest rate Lt .

[0048] S120. Determine autoregressive parameters by processing the historical bond data based on the ultimate forward rate and the Smith-Wilson model.

[0049] Among them, historical bond data also include observed interest rate data in the market, such as bond market quotations; the Smith-Wilson model is a model that applies the ultimate forward interest rate. The Smith-Wilson model is used to fit historical bond data to determine the parameters of the first-order vector autoregression process, that is, the autoregression parameters.

[0050] Specifically, the Smith-Wilson model of terminal forward rates is applied to process historical bond data to determine the autoregressive parameters.

[0051] S130. Calculate bond price simulation statistics based on the predicted bond yield obtained by processing the historical bond data with the Kalman filter.

[0052] Among them, the predicted bond yield is the predicted value y of the bond yield obtained by the Kalman filter by processing historical bond data t ;By statistically analyzing the predicted bond yields, bond price simulation statistics can be obtained. The bond price simulation statistics include the mean and variance of the simulated bond prices.

[0053] Specifically, the predicted bond yield obtained by processing historical bond data with a Kalman filter is statistically calculated to obtain bond price simulation statistics.

[0054] S140. Calculate bond yields based on the autoregressive parameters and bond price simulation statistics.

[0055] The bond yield is the bond yield that is ultimately predicted in this embodiment; the bond yield can be calculated by further processing the bond price simulation statistics through autoregressive parameters.

[0056] Specifically, the bond price simulation statistics are processed through autoregressive parameters to obtain the bond yield.

[0057] It should be noted that this embodiment determines the ultimate forward rate based on historical bond data and a preset Kalman filter; determines the autoregressive parameter based on the ultimate forward rate and the Smith-Wilson model to process the historical bond data; calculates the bond price simulation statistic based on the predicted bond yield obtained by processing the historical bond data with the Kalman filter; and calculates the bond yield based on the autoregressive parameter and the bond price simulation statistic. Through the above implementation, not only can the historical bond data be accurately fitted, but also the parameters can be accurately estimated, which is convenient for improving the prediction accuracy of bond yields.

[0058] Embodiment 2

[0059] A bond yield prediction method provided in Example 2 of the present application optimizes the "determining the ultimate forward interest rate based on historical bond data and a preset Kalman filter" in Example 1; it should be noted that for the parts not described in detail in this embodiment, reference may be made to the descriptions of other embodiments, and the method includes:

[0060] S211, determining filter parameters to be estimated of the Kalman filter.

[0061] Among them, calculate B t+1∣t The variance of the residual error of the prediction process used is Σ η It can be further written as: Q η =Σ η The decomposition factor of the filter to be estimated includes the attenuation factor λ, the variance of the residual of the observation process Σ ε , coefficient matrix Φ, mean vector μ, and decomposition factor Q η ; and the set of the above filter parameters to be estimated is recorded as ψ, ψ = {λ, Σ ε ,Φ,μ,Q η}.

[0062] S212: construct a likelihood function corresponding to the filter parameters to be estimated, and maximize the likelihood function to determine target filter parameters.

[0063] Among them, the likelihood function is used to determine the specific value of each filter parameter to be estimated; the likelihood function is a log-likelihood function corresponding to the set ψ of each filter parameter to be estimated, and the expression of the log-likelihood function is as follows:

[0064]

[0065] Wherein, T is the total number of preset expiration dates; the target filter parameter is the specific value of the corresponding filter parameter to be estimated; by maximizing the above log-likelihood function, ψ can be determined, that is, the specific value of the filter parameter to be estimated can be determined, and then the target filter parameter can be obtained.

[0066] Specifically, a log-likelihood function l(ψ) corresponding to a set ψ of each filter parameter to be estimated is established, and the log-likelihood function l(ψ) is maximized to determine ψ, that is, to determine the target filter parameter.

[0067] S213. Determine the ultimate forward interest rate based on processing historical bond data using the Kalman filter using the target filter parameters.

[0068] Among them, each target filter parameter is substituted into the Kalman filter, and then the historical bond data is processed by the Kalman filter applying the target filter parameters, that is, b t∣s With B t∣s The initialization data can be used to obtain the ultimate forward interest rate L t .

[0069] Specifically, each target filter parameter is substituted into the Kalman filter, and then the historical bond data is processed by applying the Kalman filter with the target filter parameters to obtain the ultimate forward interest rate L t .

[0070] S220. Determine autoregressive parameters by processing the historical bond data based on the ultimate forward rate and the Smith-Wilson model.

[0071] S230. Calculate bond price simulation statistics based on the predicted bond yield obtained by processing the historical bond data with the Kalman filter.

[0072] S240. Calculate bond yields based on the autoregressive parameters and bond price simulation statistics.

[0073] Embodiment 3

[0074] A bond yield prediction method provided in Example 3 of the present application optimizes the "determining the autoregressive parameters by processing the historical bond data based on the ultimate forward rate and the Smith-Wilson model" in Example 1; it should be noted that for the parts not described in detail in this embodiment, reference can be made to the descriptions of other embodiments, and the method includes:

[0075] S310. Determine the ultimate forward interest rate based on historical bond data and a preset Kalman filter.

[0076] S321. Process the historical bond data based on the ultimate forward rate and the Smith-Wilson model to obtain a model parameter matrix.

[0077] Among them, if the bond price is observed on a certain day of the maturity date, this day is recorded as the observation day d, and the calculation formula of the observation day d is as follows: in, is the discount factor. For example, if the face value of a bond is 1 yuan, but it is sold to a customer at 0.93 yuan, the corresponding discount factor is 0.93 / 1=0.93. The historical bond data includes this discount factor. q d is the vector containing the terminal forward interest rate, W d is the matrix of the equation corresponding to the Smith-Wilson model, W d The ultimate forward rate is also included in ξ; d That is, the model parameter matrix.

[0078] Specifically, the Smith-Wilson model determines the discount factor from historical bond data. And generate q based on the ultimate forward rate d With W d , further, the Smith-Wilson model will q d With W d Substitute into Calculate in order to determine the model parameter matrix ξ d .

[0079] S322. Perform principal component analysis based on the model parameter matrix to obtain target principal components.

[0080] Among them, each observation day corresponds to a corresponding model parameter matrix ξ d , the model parameter matrix ξ for the set D observation days d An overall model parameter matrix ξ can be obtained. The principal component analysis is to perform principal component analysis on the overall model parameter matrix ξ. The calculation formula for principal component analysis of ξ includes:

[0081]

[0082] Among them, when the number of principal components is controlled to be p, is a D*p matrix, i.e., the target principal component, A is a coefficient matrix consisting of p*N eigenvectors, μ is a D*N row matrix, ∈ is a D*N residual matrix; at this time, the discount factor It can be expressed as:

[0083]

[0084] It should be noted that due to the above q d With W d As time goes by, it will not change, so q d With W d Denoted as q and W respectively; μ d is the row matrix corresponding to the corresponding observation day d.

[0085] Specifically, the principal component analysis is performed on the overall model parameter matrix ξ composed of the model parameter matrix to obtain the target principal component

[0086] S323. Perform a first-order vector autoregression operation on the target principal component to determine autoregression parameters.

[0087] Among them, the first-order vector autoregression operation on the target principal component can adopt a first-order vector autoregression model, and the calculation formula for the first-order vector autoregression operation on the target principal component is as follows:

[0088]

[0089] in, For The result obtained after performing a first-order vector autoregression operation; Π and are all model parameters of the first-order vector autoregression model mentioned above, that is, autoregression parameters; is a 1×p vector, c is a 1×p vector, is a 1×p vector, Π is a p×p matrix, ∈ d is a 1×p vector.

[0090] Specifically, the target principal component is obtained by using a preset first-order vector autoregression model. Perform a first-order vector autoregression operation to determine the autoregression parameters Π and

[0091] S330. Calculate bond price simulation statistics based on the predicted bond yield obtained by processing the historical bond data with the Kalman filter.

[0092] S340. Calculate bond yields based on the autoregressive parameters and bond price simulation statistics.

[0093] Embodiment 4

[0094] A bond yield prediction method provided in Example 4 of the present application optimizes the "calculating bond price simulation statistics based on the predicted bond yield obtained by processing the historical bond data by the Kalman filter" in Example 1; it should be noted that for the parts not described in detail in this embodiment, reference can be made to the descriptions of other embodiments, and the method includes:

[0095] S410, determining the ultimate forward interest rate based on historical bond data and a preset Kalman filter.

[0096] S420. Determine autoregressive parameters by processing the historical bond data based on the ultimate forward rate and the Smith-Wilson model.

[0097] S431. Process the historical bond data based on the Kalman filter to obtain a predicted bond yield.

[0098] Among them, the Kalman filter processes historical bond data to obtain the predicted bond yield y t , with a number of maturity dates T, will predict the bond yield y t Denoted as y(t, T).

[0099] Specifically, the Kalman filter processes historical bond data to obtain the predicted bond yield y(t, T).

[0100] S432. Count the predicted bond yields to obtain target bond yield statistics.

[0101] Among them, by statistically analyzing the predicted bond yield y(t, T), the target bond yield statistics can be obtained. The target bond yield statistics include the target mean bond yield E[y(t, T)] and the target bond yield variance Var[y(t, T)].

[0102] Specifically, mathematical statistics are performed on the predicted bond yield y(t, T) to obtain the target mean of the bond yield E[y(t, T)] and the target variance of the bond yield Var[y(t, T)], thereby obtaining the target statistics of the bond yield.

[0103] S433. Calculate bond price simulation statistics based on the bond yield target statistics.

[0104] Where, the bond yield target mean E[y(t, T)] in the bond yield target statistics is recorded as a, and the bond yield target variance Var[y(t, T)] in the bond yield target statistics is recorded as b; the mean of the bond price corresponding to an observation day d is: The variance of the bond price corresponding to an observation day d is The target principal component can be Recursively iterate n steps and get Furthermore, the target principal component after recursive iteration n steps after the termination expiration date D can be calculated: The mean and variance

[0105]

[0106] Furthermore, the mean bond price after recursive iteration n steps after the termination maturity date D is E[m D+n ] and variance Var[m D+n ]for:

[0107]

[0108] The bond price simulation statistics include the mean bond price E[m D+n ] and variance Var[m D+n ].

[0109] S440. Calculate bond yields based on the autoregressive parameters and bond price simulation statistics.

[0110] Embodiment 5

[0111] A bond yield prediction method provided in Example 5 of the present application optimizes the "calculating the bond yield based on the autoregressive parameter and the bond price simulation statistic" in Example 1; it should be noted that for the parts not described in detail in this embodiment, reference can be made to the descriptions of other embodiments, and the method includes:

[0112] S510. Determine the ultimate forward interest rate based on historical bond data and a preset Kalman filter.

[0113] S520: Determine autoregressive parameters by processing the historical bond data based on the ultimate forward rate and the Smith-Wilson model.

[0114] S530: Calculate bond price simulation statistics based on the predicted bond yield obtained by processing the historical bond data with the Kalman filter.

[0115] S541. Optimize the autoregressive parameters based on a preset basin jumping algorithm to obtain target regression parameters.

[0116] Among them, the preset basin jumping algorithm is used to adjust the autoregressive parameters Π and Optimize to obtain target regression parameters corresponding to each autoregressive parameter one by one; in other embodiments, the autoregressive parameters can be optimized by basin jumping algorithm combined with “Nelder-Mead” optimization method to obtain target regression parameters corresponding to each autoregressive parameter one by one; and The corresponding target regression parameter is recorded as c * , the target regression parameter corresponding to Π is recorded as Π * , will also The corresponding target regression parameter is recorded as Σ * ;

[0117] Specifically, the preset basin jumping algorithm is used to adjust the autoregressive parameters Π and Optimize to get The corresponding target regression parameter c * , Π corresponds to the target regression parameter Π * ,as well as The corresponding target regression parameter Σ * .

[0118] S542. Calculate bond yield based on the target regression parameter and the bond price simulation statistics.

[0119] Among them, the bond yield can be calculated based on the target regression parameters and the bond price simulation statistics.

[0120] Embodiment 6

[0121] A bond yield prediction method provided in Example 6 of the present application optimizes the "calculating the bond yield based on the target regression parameter and the bond price simulation statistic" in Example 5; it should be noted that for the parts not described in detail in this embodiment, reference can be made to the descriptions of other embodiments, and the method includes:

[0122] S610. Determine the ultimate forward interest rate based on historical bond data and a preset Kalman filter.

[0123] S620: Determine autoregressive parameters by processing the historical bond data based on the ultimate forward rate and the Smith-Wilson model.

[0124] S630: Calculate bond price simulation statistics based on the predicted bond yield obtained by processing the historical bond data with the Kalman filter.

[0125] S641. Optimize the autoregressive parameters based on a preset basin jumping algorithm to obtain target regression parameters.

[0126] S642A. Calculate theoretical bond price statistics based on the target regression parameter and the bond price simulation statistics.

[0127] Among them, the target regression parameter c * ,Π * ,as well as Substituting into E[m D+n ] and Var[m D+n ] to obtain the theoretical bond price mean E * [m D+n ] and the bond theoretical price variance Var * [m D+n ];

[0128]

[0129] Among them, the bond theoretical price statistics include the bond theoretical price mean E * [m D+n ] and the theoretical bond price variance Var * [m D+n ].

[0130] Specifically, the target regression parameter c * ,Π * ,as well as Substituting into E[m D+n ] and Var[m D+n ] to obtain the theoretical bond price statistics.

[0131] S642B. Determine a simulated bond price based on a preset innovation item and the bond theoretical price statistics.

[0132] In order to obtain a bond simulation price with higher prediction accuracy, this embodiment also introduces an innovation term, thereby giving randomness to the simulation path, and the innovation term is recorded as γ is a normally distributed random quantity with mean 0 and variance 1, Σ * The bond simulation price is based on the bond theoretical price m corresponding to the innovation item and the bond theoretical price statistic. D+n Calculated; and the simulated bond price is recorded as m D+n The calculation formula is as follows:

[0133]

[0134] To enter the innovation items into m D+n The result is obtained after calculation in the calculation formula.

[0135] Specifically, determine the bond theoretical price m corresponding to the bond theoretical price statisticD+n , and then substitute the innovation term into m D+n The bond simulation price is calculated as

[0136] S642C. Calculate the bond yield based on the simulated bond price.

[0137] Among them, bond yields The calculation formula is:

[0138]

[0139] Among them, τ represents the bond simulation price The corresponding τth maturity date.

[0140] Specifically, the bond simulation price and simulated bond prices Substitute the corresponding τ into the bond yield The bond yield is calculated using the formula

[0141] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0142] Embodiment 7

[0143] Based on the same inventive concept, this embodiment also provides a bond yield prediction device for implementing the bond yield prediction method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more embodiments of the bond yield prediction device provided below can refer to the limitations of the bond yield prediction method above, and will not be repeated here.

[0144] In this embodiment, Figure 2 As shown, a bond yield prediction device is provided, comprising:

[0145] An interest rate determination module, used to determine the ultimate forward interest rate based on historical bond data and a preset Kalman filter;

[0146] A parameter determination module, used for determining autoregressive parameters by processing the historical bond data based on the ultimate forward rate and the Smith-Wilson model;

[0147] A statistics calculation module, used for calculating bond price simulation statistics based on the predicted bond yield obtained by processing the historical bond data by the Kalman filter;

[0148] The yield prediction module is used to calculate the bond yield based on the autoregressive parameter and the bond price simulation statistics.

[0149] Each module in the above bond yield prediction device can be implemented in whole or in part by software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute the operations corresponding to each module.

[0150] It should be noted that this embodiment determines the ultimate forward rate based on historical bond data and a preset Kalman filter; determines the autoregressive parameter based on the ultimate forward rate and the Smith-Wilson model to process the historical bond data; calculates the bond price simulation statistic based on the predicted bond yield obtained by processing the historical bond data with the Kalman filter; and calculates the bond yield based on the autoregressive parameter and the bond price simulation statistic. Through the above implementation, not only can the historical bond data be accurately fitted, but also the parameters can be accurately estimated, which is convenient for improving the prediction accuracy of bond yields.

[0151] In one embodiment, in determining the ultimate forward interest rate based on historical bond data and a preset Kalman filter, the interest rate determination module is specifically used to:

[0152] Determining filter parameters to be estimated of the Kalman filter;

[0153] Constructing a likelihood function corresponding to the filter parameters to be estimated, and maximizing the likelihood function to determine the target filter parameters;

[0154] The ultimate forward rate is determined based on processing historical bond data with the Kalman filter using the target filter parameters.

[0155] In one embodiment, in determining the autoregressive parameters based on the terminal forward rate and the Smith-Wilson model processing the historical bond data, the parameter determination module is specifically used to:

[0156] Processing the historical bond data based on the ultimate forward rate and the Smith-Wilson model to obtain a model parameter matrix;

[0157] Performing principal component analysis based on the model parameter matrix to obtain a target principal component;

[0158] A first-order vector autoregression operation is performed on the target principal component to determine the autoregression parameters.

[0159] In one embodiment, in calculating bond price simulation statistics based on the predicted bond yield obtained by processing the historical bond data by the Kalman filter, the statistics calculation module is specifically used to:

[0160] Processing the historical bond data based on the Kalman filter to obtain a predicted bond yield;

[0161] The predicted bond yields are counted to obtain a target bond yield statistic;

[0162] The bond price simulation statistics are calculated based on the bond yield target statistics.

[0163] In one embodiment, when calculating the bond yield based on the autoregressive parameter and the bond price simulation statistics, the yield prediction module is specifically used to:

[0164] Optimizing the autoregressive parameters based on a preset basin jumping algorithm to obtain target regression parameters;

[0165] The bond yield is calculated based on the target regression parameter and the bond price simulation statistics.

[0166] In one embodiment, in calculating the bond yield based on the target regression parameter and the bond price simulation statistic, the yield prediction module is specifically used to:

[0167] Calculate theoretical bond price statistics based on the target regression parameter and the bond price simulation statistics;

[0168] Determining a bond simulation price based on a preset innovation item and the bond theoretical price statistics;

[0169] The bond yield is calculated based on the simulated bond price.

[0170] Embodiment 8

[0171] In this embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 3As shown. The computer device includes a processor, a memory and a network interface connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a bond yield prediction method is implemented.

[0172] Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the scheme of the present disclosure, and does not constitute a limitation on the computer device to which the scheme of the present disclosure is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0173] Embodiment 9

[0174] In this embodiment, a computer readable storage medium is provided. Figure 4 As shown, a computer program is stored thereon, and when the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0175] Embodiment 10

[0176] In this embodiment, a computer program product is provided, including a computer program. When the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0177] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this disclosure are all information and data authorized by the user or fully authorized by all parties.

[0178] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided by the present disclosure can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided by the present disclosure may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited thereto. The processor involved in each embodiment provided by the present disclosure may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited thereto.

[0179] The technical features of the above embodiments may be arbitrarily combined. 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.

[0180] The above-described embodiments only express several implementation methods of the present disclosure, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present invention. It should be pointed out that, for a person of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present disclosure, and these all belong to the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure shall be subject to the attached claims.

Claims

1. A bond yield prediction method, characterized in that: include: Determine the ultimate forward rate based on historical bond data and a preset Kalman filter; Determine autoregressive parameters by processing the historical bond data based on the terminal forward rate and the Smith-Wilson model; Calculating bond price simulation statistics based on the predicted bond yield obtained by processing the historical bond data with the Kalman filter; The bond yield is calculated based on the autoregressive parameters and the bond price simulation statistics.

2. The method according to claim 1, characterized in that The method of determining the ultimate forward interest rate based on historical bond data and a preset Kalman filter includes: Determining filter parameters to be estimated of the Kalman filter; Constructing a likelihood function corresponding to the filter parameters to be estimated, and maximizing the likelihood function to determine the target filter parameters; The ultimate forward rate is determined based on processing historical bond data with the Kalman filter using the target filter parameters.

3. The method according to claim 1, characterized in that The determining of the autoregressive parameters by processing the historical bond data based on the ultimate forward rate and the Smith-Wilson model includes: Processing the historical bond data based on the ultimate forward rate and the Smith-Wilson model to obtain a model parameter matrix; Performing principal component analysis based on the model parameter matrix to obtain a target principal component; A first-order vector autoregression operation is performed on the target principal component to determine the autoregression parameters.

4. The method according to claim 1, characterized in that The method of calculating bond price simulation statistics based on the predicted bond yield obtained by processing the historical bond data with the Kalman filter includes: Processing the historical bond data based on the Kalman filter to obtain a predicted bond yield; The predicted bond yields are counted to obtain a target bond yield statistic; The bond price simulation statistics are calculated based on the bond yield target statistics.

5. The method according to claim 1, characterized in that The calculating of bond yield based on the autoregressive parameter and bond price simulation statistics includes: Optimizing the autoregressive parameters based on a preset basin jumping algorithm to obtain target regression parameters; The bond yield is calculated based on the target regression parameter and the bond price simulation statistics.

6. The method according to claim 5, characterized in that The calculating of the bond yield based on the target regression parameter and the bond price simulation statistic includes: Calculate theoretical bond price statistics based on the target regression parameter and the bond price simulation statistics; Determining a bond simulation price based on a preset innovation item and the bond theoretical price statistics; The bond yield is calculated based on the simulated bond price.

7. A bond yield prediction device, characterized in that: The device comprises: An interest rate determination module, used to determine the ultimate forward interest rate based on historical bond data and a preset Kalman filter; A parameter determination module, used for determining autoregressive parameters by processing the historical bond data based on the ultimate forward rate and the Smith-Wilson model; A statistics calculation module, used for calculating bond price simulation statistics based on the predicted bond yield obtained by processing the historical bond data by the Kalman filter; The yield prediction module is used to calculate the bond yield based on the autoregressive parameter and the bond price simulation statistics.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images