WO2021103571A1 - Method and apapratus for generating asset investment suggestion information and readable storage medium - Google Patents

Abstract

A method and apparatus for generating asset investment suggestion information and a readable storage medium. The method for generating asset investment suggestion information comprises the following steps: acquiring various assets so as to determine year-on-year sequential data corresponding to the various assets, the classification means for the various assets comprising asset type classification and investment risk classification (S100); determining, according to the various year-on-year sequential data, a predicted fluctuation amplitude corresponding to each asset (S200); and determining, according to the various predicted fluctuation amplitudes, the investment weights of the various assets (S300).

Description

Method, device and readable storage medium for generating asset investment suggestion information

Priority information

This application claims the priority of the Chinese patent application filed on November 25, 2019, the application number is 201911169350.X, and the name is "Method, device and readable storage medium for generating asset investment advice information", all of which are approved The reference is incorporated in this application.

Technical field

This application relates to the field of computer technology, and in particular to a method, device and readable storage medium for generating asset investment advice information.

Background technique

With the improvement of people's living standards, the liquidity in their hands becomes more and more surplus. Compared with depositing funds in banks, people are more willing to invest their funds in high-yielding financial products such as stocks, funds, and bonds.

In the prior art, when investing in wealth management products, people look at the description of the risks and returns of the wealth management products, and invest funds based on their own investment experience and the risks and returns of the wealth management products. It can be seen that the user's investment method is determined by the user's investment experience and the expected return of financial products. This investment method is less reliable and cannot bring users a stable and reliable investment return.

The above content is only used to assist the understanding of the technical solution of the application, and does not mean that the above content is recognized as prior art.

Summary of the invention

The main purpose of this application is to provide a method, device and readable storage medium for generating asset investment advice information, aiming to solve the problem of poor reliability of capital investment methods in the prior art.

In order to achieve the above-mentioned purpose, this application provides a method for generating asset investment advice information. The method for generating asset investment advice information includes the following steps:

Obtain various types of assets to determine the year-on-year sequence data corresponding to each type of said assets, where the classification methods of each type of said assets include asset type classification and investment risk classification,

Determine the predicted increase and decrease corresponding to each type of asset according to each of the said year-on-year sequence data;

Determine the investment weights of the various types of assets according to each of the predicted fluctuations.

In an embodiment, the step of determining the predicted fluctuation range corresponding to each type of the asset according to each of the year-on-year sequence data includes:

Determine the filter sequence corresponding to each type of asset in each preset observation period according to the year-on-year sequence data;

Combine the filter sequences corresponding to the various types of assets to obtain the combined sequence of each of the preset observation periods, and determine the output length of the combined sequence, so as to modify the output length by the preset length to obtain Extrapolate the combined sequence;

Input each of the year-on-year sequence data and the combined sequence into a linear regression model to obtain linear regression coefficients;

According to the extrapolated combined sequence and the linear regression coefficient, a fitting prediction of the same-year sequence data is performed on each type of asset to determine the predicted increase or decrease corresponding to the various types of assets.

In an embodiment, the step of combining the filter sequences corresponding to the various types of assets to obtain the combined sequence of each of the preset observation periods includes:

Determine investment risk parameters corresponding to various types of assets, and classify the various types of assets according to the investment risk parameters to obtain various sets, and the investment risk parameters corresponding to the various types of assets in the set belong to the same numerical range;

Merging the filter sequences corresponding to the various types of assets in the collection to obtain an intermediate merging sequence of each of the collections in each of the preset observation periods;

Merging each of the intermediate merged sequences to obtain a merged sequence corresponding to each of the preset observation periods.

In an embodiment, the step of inputting each of the year-on-year sequence data and the combined sequence into a linear regression model to obtain linear regression coefficients includes:

Each of the year-on-year sequence data is used as a dependent variable, and the combined sequence is used as an independent variable to input into a linear regression model to obtain a linear regression coefficient.

In an embodiment, the step of determining the investment weights corresponding to the various types of assets according to each of the predicted fluctuations includes:

Sort each of the predicted fluctuations in descending order;

Determine the investment weight of the asset corresponding to each of the predicted fluctuations according to each of the predicted fluctuations in the order, wherein, the greater the predicted fluctuation, the greater the investment weight of the asset corresponding to the predicted fluctuation Big.

In one embodiment, according to the extrapolated combined sequence and the linear regression coefficient, the fitting prediction of the same-year sequence data is performed on each type of asset, so as to determine the predicted fluctuation rate corresponding to each type of asset The steps include:

According to the extrapolated combined sequence and the linear regression coefficient, perform a fitting prediction of the same-year sequence data for each type of asset to determine the predicted value corresponding to the various types of assets;

Determine the fitting values corresponding to the various types of assets at the current time;

The predicted value and the fitted value corresponding to each type of asset are differentiated to obtain the predicted increase or decrease corresponding to each type of asset.

In an embodiment, the step of determining the filter sequence corresponding to each type of asset in each preset observation period according to the year-on-year sequence data includes:

Performing zero padding on the year-on-year sequence data, and performing Fourier transform on the year-on-year sequence data after zero padding to obtain corresponding frequency domain data;

Determine a set of filter coefficients according to each of the preset observation periods, and obtain an intermediate sequence according to the filter coefficients and the frequency domain data;

Perform an inverse Fourier transform on the intermediate sequence to obtain a filter sequence of the asset in each of the preset observation periods.

In order to achieve the above objective, the present application also provides a device for generating asset investment advice information. The device for generating asset investment advice information includes: a memory, a processor, and a memory, a processor, and a memory that is stored on the memory and can run on the processor. A program for generating asset investment recommendation information, which implements the steps of the method for generating asset investment recommendation information as described above when the program for generating asset investment recommendation information is executed by the processor.

In order to achieve the above object, the present application also provides a readable storage medium, the readable storage medium stores an asset investment advice information generation program, and the asset investment advice information generation program is executed by a processor to achieve the above The steps of the method for generating asset investment advice information described above.

The method, device and readable storage medium for generating asset investment recommendation information proposed by the embodiments of the application. The device obtains various types of assets and the corresponding year-on-year sequence data of various types of assets, and determines the corresponding assets of each type according to the same-year sequence data. Forecast the rise and fall, and then determine the investment weight of various assets based on each predicted rise and fall. Because the device can determine the asset type classification or the investment weight of various assets after the investment risk classification through the year-on-year sequence data of various assets, the device can generate investment advice information for various assets according to the investment weight reasonably, and then provide users with reliable Investment advice for wealth management products with stable income.

Description of the drawings

FIG. 1 is a schematic diagram of the hardware structure of the device for generating asset investment recommendation information involved in the scheme of the embodiment of the present application;

FIG. 2 is a schematic flowchart of a first embodiment of a method for generating investment advice information for assets in an application;

FIG. 3 is a schematic flowchart of a second embodiment of a method for generating investment advice information of an application asset;

FIG. 4 is a schematic flowchart of a third embodiment of a method for generating investment advice information for an application for assets;

Fig. 5 is a schematic flowchart of a fourth embodiment of a method for generating asset investment advice information for the application.

The realization, functional characteristics, and advantages of the purpose of this application will be further described in conjunction with the embodiments and with reference to the accompanying drawings.

Detailed ways

It should be understood that the specific embodiments described here are only used to explain the present application, and are not used to limit the present application.

The main solution of the embodiment of this application is to obtain various types of assets to determine the year-on-year sequence data corresponding to the various types of assets. The classification methods of the various types of assets include asset type classification and investment risk classification. The year-on-year sequence data determines the predicted rise and fall of each type of asset; the investment weight of each category of the asset is determined according to each of the predicted rises and falls.

Because the device can determine the asset type classification or the investment weight of various assets after the investment risk classification through the year-on-year sequence data of various assets, the device can generate investment advice information for various assets according to the investment weight reasonably, and then provide users with reliable Investment advice for wealth management products with stable income.

As shown in Fig. 1, Fig. 1 is a schematic diagram of the hardware structure of the device for generating asset investment recommendation information involved in the solution of the embodiment of the present application.

As shown in FIG. 1, the device for generating asset investment recommendation information may include: a processor 1001, such as a CPU, a communication bus 1002, and a memory 1003. Among them, the communication bus 1002 is used to implement connection and communication between these components. The memory 1003 may be a high-speed RAM memory, or a non-volatile memory (non-volatile memory), such as a magnetic disk memory. Optionally, the memory 1003 may also be a storage device independent of the aforementioned processor 1001.

Those skilled in the art can understand that the structure of the device for generating asset investment advice information shown in FIG. 1 does not constitute a limitation on the device for generating asset investment advice information, and may include more or less components than shown in the figure, or a combination Certain components, or different component arrangements.

As shown in FIG. 1, the memory 1003, which is a computer storage medium, may include an operating system and a program for generating asset investment advice information.

In the device shown in FIG. 1, the processor 1001 may be used to call the asset investment recommendation information generation program stored in the memory 1003, and perform the following operations:

Obtain various types of assets to determine the year-on-year sequence data corresponding to the various types of assets, and determine the filter sequence corresponding to each type of asset in each preset observation period according to the year-on-year sequence data;

Combining the filter sequences corresponding to the various types of assets to obtain the combined sequence of each of the preset observation periods;

According to the combined sequence and each of the year-on-year sequence data, the investment weights of the various types of assets are determined.

In an embodiment, the processor 1001 may call a processing program for generating asset investment recommendation information stored in the memory 1003, and also perform the following operations:

Obtain various types of assets to determine the year-on-year sequence data corresponding to each type of said assets, where the classification methods of each type of said assets include asset type classification and investment risk classification,

Determine the predicted increase and decrease corresponding to each type of asset according to each of the said year-on-year sequence data;

Determine the investment weights of the various types of assets according to each of the predicted fluctuations.

In an embodiment, the processor 1001 may call a processing program for generating asset investment recommendation information stored in the memory 1003, and also perform the following operations:

Determine the filter sequence corresponding to each type of asset in each preset observation period according to the year-on-year sequence data;

Combine the filter sequences corresponding to the various types of assets to obtain the combined sequence of each of the preset observation periods, and determine the output length of the combined sequence, so as to modify the output length by the preset length to obtain Extrapolate the combined sequence;

Input each of the year-on-year sequence data and the combined sequence into a linear regression model to obtain linear regression coefficients;

According to the extrapolated combined sequence and the linear regression coefficient, perform a fitting prediction of the year-on-year sequence data for each type of asset to determine the predicted increase or decrease corresponding to the various types of assets;

Determine the investment weights corresponding to the various types of assets according to each of the predicted fluctuations.

In an embodiment, the processor 1001 may call a processing program for generating asset investment recommendation information stored in the memory 1003, and also perform the following operations:

Determine investment risk parameters corresponding to various types of assets, and classify the various types of assets according to the investment risk parameters to obtain various sets, and the investment risk parameters corresponding to the various types of assets in the set belong to the same numerical range;

Merging the filter sequences corresponding to the various types of assets in the collection to obtain an intermediate merging sequence of each of the collections in each of the preset observation periods;

Merging each of the intermediate merged sequences to obtain a merged sequence corresponding to each of the preset observation periods.

In an embodiment, the processor 1001 may call a processing program for generating asset investment recommendation information stored in the memory 1003, and also perform the following operations:

Each of the year-on-year sequence data is used as a dependent variable, and the combined sequence is used as an independent variable to input into a linear regression model to obtain a linear regression coefficient.

In an embodiment, the processor 1001 may call a processing program for generating asset investment recommendation information stored in the memory 1003, and also perform the following operations:

Sort each of the predicted fluctuations in descending order;

Determine the investment weight of the asset corresponding to each of the predicted fluctuations according to each of the predicted fluctuations in the order, wherein, the greater the predicted fluctuation, the greater the investment weight of the asset corresponding to the predicted fluctuation Big.

In an embodiment, the processor 1001 may call a processing program for generating asset investment recommendation information stored in the memory 1003, and also perform the following operations:

According to the extrapolated combined sequence and the linear regression coefficient, perform a fitting prediction of the same-year sequence data for each type of asset to determine the predicted value corresponding to the various types of assets;

Determine the fitting values corresponding to the various types of assets at the current time;

The predicted value and the fitted value corresponding to each type of asset are differentiated to obtain the predicted increase or decrease corresponding to each type of asset.

In an embodiment, the processor 1001 may call a processing program for generating asset investment recommendation information stored in the memory 1003, and also perform the following operations:

Performing zero padding on the year-on-year sequence data, and performing Fourier transform on the year-on-year sequence data after zero padding to obtain corresponding frequency domain data;

Determine a set of filter coefficients according to each of the preset observation periods, and obtain an intermediate sequence according to the filter coefficients and the frequency domain data;

Perform an inverse Fourier transform on the intermediate sequence to obtain a filter sequence of the asset in each of the preset observation periods.

Based on the above hardware construction, various embodiments of the method for generating asset investment recommendation information of the present application are proposed.

2, the first embodiment of the present application provides a method for generating asset investment advice information, the method includes:

Step S100, acquiring various types of assets to determine the year-on-year sequence data corresponding to the various types of assets;

In this embodiment, the execution subject is the device for generating asset investment recommendation information. For ease of description, the device below refers to the device for generating asset investment recommendation information. The device can be regarded as a server, and the device can be loaded into the client in the form of an APP, so that the client communicates with the device based on the APP.

In this embodiment, the assets can be funds, bonds, stocks, commodities, and so on. The user can open the APP loaded on the client. The APP is the investment program of the wealth management product; the user can select the wealth management product that needs to be invested based on the APP. For example, if the user wants to invest in A stocks and B bonds, the user can select the interface based on the assets of the APP Select A stocks and B bonds on the app, and the APP sends the A stocks and B bonds selected by the user to the device, so that the device can obtain various assets based on the A stocks and B bonds.

Specifically, after the device obtains the wealth management product selected by the user, the wealth management product needs to be classified into assets to obtain various assets. The method of asset classification can be classified according to the type of asset. For example, wealth management products can be classified into stocks, bonds, commodities, etc.; the method of asset classification can also be classified according to the level of investment risk, for example, the risk of investment can be higher. Stocks and commodities are regarded as one type of assets, and bonds with lower investment risks are regarded as another type of assets. It is understandable that the classification methods include asset type classification and investment risk classification.

After acquiring various types of assets, the device needs to determine the corresponding year-on-year sequence data for each type of asset. The data included in the year-on-year series data are all monthly year-on-year data. Each month’s year-on-year data is calculated in a fixed 12-month cycle. For example, if the asset is a stock, the ratio of the closing price of the stock at the end of September 2019 to the closing price of the stock at the end of September 2018 or the logarithm of the ratio is taken as 2019 The year-on-year serial data in September of 2009, and September 2019 is the time of observation. The year-on-year serial data

Calculated according to the following formula:

t ₀ = September 2018, t ₀ +12 = September 2019.

The period of monthly year-on-year data is 12 months.

Step S200: Determine the predicted increase or decrease corresponding to each type of asset according to each of the year-on-year sequence data;

After determining each year-on-year sequence data, the device obtains the predicted rise and fall of each type of asset based on each year-on-year sequence data. Specifically, referring to FIG. 3, that is, step S200 includes:

Step S210: Determine the filter sequence corresponding to each type of asset in each preset observation period according to the year-on-year sequence data;

After determining the year-on-year sequence data of various types of assets, the device then determines the filter sequence corresponding to each type of asset in each preset observation period according to the year-on-year sequence data.

Specifically, the asset data sequence in a continuous period of time has a time dimension. For example, the stock price data is a numerical sequence in continuous time, so the resulting year-on-year sequence data of assets can be regarded as a time sequence. The time series can be analogous to the time-domain signal generated by the movement of an economic and financial asset, and the time-domain signal has the expression of Fourier series. Therefore, Fourier transform can be performed on the year-on-year sequence data to obtain the corresponding frequency. Data in the frequency domain is analyzed and processed in the frequency domain.

Different types of assets have a common business cycle in a unified financial and economic system. The business cycle is the preset observation cycle. The year-on-year data of assets show regular changes in the preset observation cycle, and the preset observation cycle is 42 Months, 100 months and 200 months, the preset observation period is set in the device. In the frequency domain, each preset observation period corresponds to a target frequency signal. These target frequency signals are stable and continuous, while other unstable or unsustainable frequency signals can be regarded as noise. Therefore, the device performs frequency domain filtering on the year-on-year sequence data of each type of asset to retain the target frequency signal and reduce noise interference. The year-on-year sequence data of the reserved target frequency signal is the frequency domain data.

Due to the fence effect in the Fourier transform, when obtaining the filter sequence of each type of asset, first add zeros to the year-on-year sequence data, and perform Fourier transform on the zero-padded year-on-year sequence data to obtain the corresponding frequency. Domain data. To

Represents the year-on-year serial data of the i-th asset class, with

Represents the frequency domain data corresponding to the year-on-year sequence data of the i-th asset.

As shown in the following formulas (1), (2), (3), a set of filter coefficients gauss _{win is} determined according to each preset observation period period, and according to each set of filter coefficients gauss _win and frequency domain data wave _fft Get a set of intermediate sequences

Among them, nfft is the zero-padded length of the year-on-year data, the gauss _index is a series of numbers from 1 to nfft, center _frequency represents the center frequency, that is, the frequency corresponding to the period factor to be extracted, and gauss _alpha is a parameter that affects the Gaussian filter bandwidth. These parameters can be preferably set as follows: nfft takes 4096, period takes the corresponding 42 months, 100 months or 200 months, and gauss _alpha takes 10.

It should be noted that according to the following formula (4)

Perform conjugate symmetry operations:

Then the intermediate sequence of each preset observation period period in the i-th asset

Perform inverse Fourier transform to obtain the filtering sequence of the i-th asset in each preset observation period

The formula (5) is obtained through the transformation of the above formula. The transformation process is: perform inverse Fourier transform on each set of intermediate sequences to obtain a set of second intermediate sequences; intercept data from the second intermediate sequences according to the preset sequence length LEN Click to get the filtering sequence of each type of asset under each preset observation period

Among them, the preset sequence length LEN is equal to the sum of the corresponding sequence length L1 and the extrapolated length L2.

Among them, Real(Z) is the real part of Z.

Step S220: Combine the filter sequences corresponding to the various types of assets to obtain the combined sequence of each of the preset observation periods, and determine the output length of the combined sequence to perform a preset length of the output length Modify to get the extrapolated combined sequence;

In the face of the same global economic and financial environment, multiple types of assets are driven by the same economic cycle and exhibit extremely relevant behaviors. Therefore, for each type of asset, it is necessary to merge the filter sequences of multiple types of assets under the same preset observation period, that is, to merge their similar common economic cycle change characteristics, and the merged sequence can reflect the unified system in the market. The level of periodic movement is used for subsequent processing to better fit the year-on-year sequence data of various asset prices.

When the merging method is the same type of asset merging method, the merging sequence of the preset observation period can be synthesized by referring to the following process:

Step A: Obtain a first filter matrix according to the filter sequence of the various types of assets;

In this embodiment, various types of assets can be stocks, bonds, and commodities, and the filter sequences of stocks, bonds, and commodities are

Taking each filter sequence as a vector, the three vectors are combined to form a filter matrix M1, and M1 is the first filter matrix.

Step B, Hilbert transform is performed on the first filter matrix to obtain a corresponding second filter matrix;

In this step, as shown in the following formula, call the software platform library function hibert to perform Hilbert transform on M1 to obtain the second filter matrix M2:

M2=hilbert (M1)

Step C: Perform iterative calculation of merging weights according to the second filter matrix.

Specifically, first initialize the combined weight vector with the preset observation period as the period

It is a vector of all 1s with length N, where N is the number of vectors in matrix M2, that is, the number of asset categories, and the value is 3 when the asset only includes stocks, bonds, and commodities.

Combine weight vectors according to the following formulas (6), (7), (8)

The iterative calculation:

weight _m = mean(weight) (7)

among them,

Represents the combined weight vector after the kth iteration calculation, M*N represents the multiplication of the matrix M and N, (M)′ is the conjugate transpose of M, and diag(W) contains W on the main diagonal Diagonal matrix, conj(M) is the complex conjugate of M, M.*N represents the dot product of matrix M and N, mean(M) is the mean value of each column of M.

When the number of iteration calculations reaches the preset iteration number threshold dcnt, the merge weights converge, and the final merge weight vector is obtained

In this implementation, dcnt=100 can be selected.

Since the filter sequences of various assets can be regarded as time-domain signals, they are affected by the cycle of the entire economic and financial system and are similar to the principles of signal propagation. Most of them are interfered by strong noise, and the signal-to-noise ratio is usually not high. The iterative calculation method of merging weights given in the step reduces the phase difference estimation error and obtains the optimal weights of the merging sequence of various assets, thereby effectively improving the signal-to-noise ratio and stability of the merging sequence of the filter sequence of various assets .

Step D: Obtain the combined sequence according to the second filter matrix and the combined weight vector.

According to the following formula (9), the synthesized sequence X _period whose preset observation period takes the value of period is a vector, namely

abs(W) is the complex magnitude of each element of W (each element of W is a complex number), and sum(W) is the sum of the elements of W.

It should be noted that the classification methods of various assets in the above examples are classified according to asset types; when the classification of assets is classified as investment risk, each asset that needs to be classified as one type of investment risk needs to be filtered first. The merger of various assets is performed again in the merger sequence of various assets to obtain the final merger sequence under the preset observation period. For example, according to investment risk, the stocks and commodities are divided into one category, and the other is bonds. The preset observation period is 42 months. First, according to the filter sequence corresponding to the stocks and commodities in 42 months, the stocks and commodities are determined The 42-month first merging sequence is combined again with the first merging sequence and the bond's 42-month filtering sequence to obtain the 42-month merging sequence of the two types of assets.

After determining the combined sequence corresponding to each preset observation period, the device needs to perform Gaussian filtering extrapolation on the combined sequence to obtain an extrapolated combined sequence.

Specifically, the combined sequence is obtained, the output length of the combined sequence is determined, and the output length is modified by a preset length, thereby obtaining the extrapolated combined sequence. For example, the output length of a 42-month merged sequence is 120, and the extrapolated filter one period will get an extrapolated merged sequence with an output length of 121. The extrapolated filter one period is the preset length, and the preset length can be based on actual needs. The setting is not limited to 1.

Step S230, input each of the year-on-year sequence data and the combined sequence into a linear regression model to obtain a linear regression coefficient;

The linear regression model is stored in the device, and the device can use the synchronized sequence data corresponding to various assets as the dependent variable, and the combined sequence as the independent variable, which is input into the linear regression model to obtain the linear regression coefficient. Specifically, the linear regression model has a corresponding formula, that is, the linear regression formula, and the linear regression formula is:

Among them, X ₄₂ , X ₁₀₀ , and X ₂₀₀ are the merged sequences corresponding to 42 months, 100 months, and 200 months without predicting the preset observation period, b ₁ is the intercept term, and b ₂ , b ₃ , and b ₄ are linear Regression coefficients,

It is the year-on-year serial data. The linear regression model uses the least squares estimation algorithm to obtain the estimated value of the linear regression coefficient according to the linear regression formula.

It should be noted that the classification of various assets can be classified according to asset category or investment risk. If the asset classification method is asset category classification, the above formula is used to obtain the linear regression coefficient; if the asset classification method is investment risk classification , You need to differentiate the year-on-year sequence data of various assets to obtain the differential data, and then use the differential data as the dependent variable and the combined sequence as the independent variable to substitute the linear regression coefficient to obtain the linear regression coefficient.

The difference data is obtained by the difference of the year-on-year serial data, as follows:

Using the linear regression equation, the fitted value LP _t (LP _t ＝b ₁ +b ₂ *X42 _filter [120]+b ₃ *X100 _filter [120]+b ₄ *X200 _filter [ 120]) and the predicted value LP _t+1 (LP _t+1 ＝b ₁ +b ₂ *X42 _filter [121]+b ₃ *X100 _filter [121]+b ₄ * X200 _filter [121]). Furthermore, the difference between the predicted value in period t+1 and the fitted value in period t is used to obtain the first-order difference of the logarithmic year-on-year sequence: ΔLP _t+1 =LP _t+1 -LP _t , ΔLP _{t+1 is} the difference data.

Step S240, according to the extrapolated and merged sequence and the linear regression coefficient, perform a fitting prediction of the year-on-year sequence data for each type of asset to determine the predicted increase or decrease corresponding to each type of the asset;

After determining the linear regression coefficients, the device can perform a fitting prediction of the same-year sequence data for each type of asset according to the extrapolated combined sequence and the linear regression coefficients to obtain the predicted value.

The fitted value at the current time is

So as to get the predicted rise and fall

In step S300, the investment weights of the various types of assets are determined according to each of the predicted fluctuations.

After determining the predicted rise and fall of various assets, the target weights of various assets can be determined according to the predicted rise and fall. Specifically, the predicted fluctuation can be positive or negative. When the predicted fluctuation is greater than 0, it indicates that the asset corresponding to the predicted fluctuation will have greater returns in the next period, and when the predicted fluctuation is less than 0 , It indicates that the assets corresponding to the predicted rise and fall will have a lower return in the next period.

The device can determine the investment weight of various assets according to the magnitude of the predicted rise and fall of various assets. The greater the predicted rise and fall, the greater the investment weight.

After determining the investment weights of various differences, the device can generate investment advice information for various assets. For example, if various types of assets are stocks, commodities, and bonds, and the corresponding weights of stocks, commodities, and bonds are 0.2, 0.5, and 0.3, it is recommended to divide the capital into 20%, 50%, and 30%. Invest funds in stocks, invest 50% of funds in commodities, and invest 30% of funds in bonds". In addition, there are multiple assets in a certain type of asset. For example, if Class A assets include stocks and bonds, the investment weight of stocks and bonds is half of the investment weight of Class A assets.

In addition, the device can also modify the investment recommendation information according to the desired wealth management product set by the user as a reference. For example, if the desired wealth management product set by the user is a medium-low risk and stable product, while the bond risk is low and stable, it is recommended to divide the funds into 20%, 50%, and 30%, and invest 20% of the funds in stocks, 50% of funds will be invested in commodities and 30% of funds will be invested in bonds", amended to "it is recommended to divide funds into 15%, 45%, and 40%, 15% of funds will be invested in stocks, 45% of funds will be invested in commodities, and Invest 40% of funds in bonds".

After the device generates the investment suggestion information, the investment suggestion information is fed back to the user's corresponding client, so that various financial products can be invested based on the investment suggestion information.

In the technical solution provided in this embodiment, the device obtains various types of assets and year-on-year sequence data corresponding to each type of asset, and determines the predicted fluctuations corresponding to each type of asset based on the year-on-year sequence data, and then determines the predicted fluctuations according to the respective predicted fluctuations. Investment weight of various assets. Since the device can determine the asset type classification or the investment weight of various assets after the investment risk classification through the year-on-year sequence data of various assets, the device can generate investment advice information of various assets according to the investment weight reasonably, and then provide users with reliable Investment advice for wealth management products with stable income.

In an embodiment, the combined sequence can also be synthesized according to the SUMPLE algorithm. Specifically, each year-on-year sequence data can be expressed as:

Where k is the time variable,

Is the data of the i-th year-on-year data at time k,

Is noise. The synthesized weight coefficient is expressed as:

Where K is the time variable in the unit of the relevant time interval n _cor , that is, the number of iterations in the synthesis,

Is the ideal weight,

Is the weight estimation error caused by noise, then the synthesized synthesized sequence can be expressed as:

Where * is the complex conjugate, and L is the total number of year-on-year sequence data. If the output of the synthesized sequence is expressed as the following form:

Then, the signal and noise terms are:

The K+1th weight coefficient in the SUMPLE algorithm

By the Kth

Recursively get:

In the formula, R _K+1 is the normalization coefficient, which can prevent the weight range from becoming unstable due to continuous accumulation. It ensures that the square sum of the weight coefficient of each year-on-year serial data is equal to the number of year-on-year serial data, namely

The weight of the above formula

_{C k} can also be rewritten as:

In this embodiment, when the SUMPLE algorithm is used to synthesize each year-on-year sequence data, the weight coefficient of each year-on-year sequence data is converged through a preset number of iterations, that is, the number of iterations for the weight coefficient convergence is determined in advance through experiments and stored as The preset number of times, when the SUMPLE algorithm is used to synthesize each year-on-year sequence data, when the number of iterations reaches the preset number of times, the iteration is stopped, and the synthesized combined sequence is output. The system-level financial data movement law represented by the merged sequence is more stable and reliable, and more predictable. At the same time, because the SUMPLE algorithm is suitable for the synthesis of low signal-to-noise ratio data, when the SUMPLE algorithm is used to synthesize each year-on-year sequence data, the optimal weight of each year-on-year sequence data can be calculated, so that the synthesized combined sequence signal noise Than higher.

In this embodiment, when the SUMPLE algorithm is used to synthesize each year-on-year sequence data, two iterative methods can be selected, either the overall iteration or the rolling iteration, to make the weight coefficients of each year-on-year sequence data converge. When synthesizing through the overall iterative method, take a sample, and then use the overall data to iteratively update itself. At this time, the relevant time interval n _cor is the duration corresponding to the overall data; when synthesizing through the rolling iteration method, The sampling window is scrolled forward, sampling multiple times, and the sequence obtained at the next time is used to update the weight coefficient obtained at the previous time. At this time, the relevant time interval n _cor is the preset duration, and the preset duration can be set according to the actual situation For example, the preset duration can be set to 4 months (or 120 days). In actual synthesis, due to the limited data length, when using rolling iterations, the number of iterations is limited by the total duration of the sampled data. Therefore, the number of iterations may be less due to insufficient data length, which affects the convergence of the weight coefficients. The number of iterations of the iterative method is not limited by the length of the same-year sequence data. Therefore, in this embodiment, preferably, when the SUMPLE algorithm is used to synthesize each year-on-year sequence data, the overall iterative method is selected for synthesis, which can ensure that the weight coefficients of each year-on-year sequence data converge.

Referring to Figure 4, Figure 4 is a second embodiment of the method for generating asset investment advice information of the application. Based on the first embodiment, acquiring various assets in step S100 to determine the year-on-year sequence data corresponding to each category of assets includes :

Step S110: Obtain the observation time, the preset lag period, and the preset year-on-year sequence length;

Research has found that the original asset price lags its year-on-year serial data, and there is a lag period. For example, the original asset price is about 5 months slower than the year-on-year serial data, that is, the change in the cycle phase will be reflected in the asset return after 5 months. Rate. In other words, for the observation time t, if

It is believed that the corresponding asset return rate will rise at the predicted time t+5, and vice versa if

It is believed that the return on assets corresponding to period t+5 will decrease.

Therefore, in this embodiment, after obtaining the observation time and the preset year-on-year sequence length, it is also necessary to obtain a preset lag period. For example, set the preset lag period to 5 months.

Step S120: Obtain an end time according to the observation time and the preset hysteresis period, and obtain a start time according to the end time and the preset year-on-year sequence length;

Step S130: Acquire year-on-year sequence data of various types of the assets from the start time to the end time.

For example, if the preset year-on-year sequence length is 120, for the observation time t = May 31, 2010, the required year-on-year sequence data is [t-(120+5)month, t-5month], that is, the date starts from 2000 The monthly sequence of stock prices starting on January 31, 2009 and ending on December 31, 2009, is used to predict the stock price trend on May 31, 2010.

In this embodiment, when predicting the asset trend, the year-on-year sequence data of multiple types of assets are obtained according to the lag period of the original asset price relative to the same-year sequence data, which reduces the error of predicting the trend of each type of asset.

Referring to FIG. 5, FIG. 5 is a third embodiment of the method for generating asset investment advice information of this application. Based on any one of the first to third embodiments, the filtering corresponding to each type of asset is performed in step S220. The sequence is combined to obtain the combined sequence of each of the preset observation periods, including:

Step S221: Determine investment risk parameters corresponding to various types of assets, and classify the various types of assets according to the investment risk parameters to obtain various sets, and the investment risk parameters corresponding to the various types of assets in the set belong to the same value. Interval

Step S222: Combine the filter sequences corresponding to the various types of assets in the collection to obtain an intermediate combined sequence of each of the collections in each of the preset observation periods;

Step S223: Combine each of the intermediate combined sequences to obtain a combined sequence corresponding to each of the preset observation periods.

In this embodiment, when the asset classification method is investment risk classification, each asset that needs to be classified as investment risk is first merged with a filter sequence, and the merge sequence of various assets is merged again to obtain The final combined sequence under the preset observation period.

Specifically, the device first determines the filter sequence corresponding to each type of asset, and the classification method of each type of asset is asset type classification. The device then determines the investment risk parameter corresponding to each type of asset. The investment risk parameter represents the investment risk of the asset. The higher the investment risk parameter, the greater the investment risk of the asset. The device can classify various types of assets according to the investment risk parameters of each type of asset, and obtain each set. The investment risk parameters of various assets in the set belong to the same numerical range. For example, there are two sets. The various assets in one set are stocks and commodities. The value range of investment risk parameters corresponding to stocks and commodities corresponds to high-risk investments; the assets in the other set are bonds, and bonds correspond to investments. The numerical range of risk parameters corresponds to low-risk investments. The preset observation period is 42 months. First, according to the 42-month filtering sequence of stocks and commodities, determine the intermediate merging sequence of stocks and commodities corresponding to 42 months, and use the bond filtering sequence as the middle of another set. Combine the sequence, and then merge the intermediate merged sequences corresponding to the two sets again at 42 months to obtain the merged sequence corresponding to the high-risk investment set and the low-risk investment set at 42 months.

In this application, the method of generating asset investment advice information can be divided into eight steps, specifically:

Step 1: Determine the year-on-year sequence data of various types of assets. The classification methods of various types of assets include asset type classification and investment risk classification. Therefore, step 1 can be divided into two situations, one of which is derived from asset type classification The year-on-year sequence data of various types of assets is defined as step 1.1; the other is the year-on-year sequence data of various assets obtained by the investment risk classification method, which is defined as step 1.2;

Step 2: Determine the filter sequence corresponding to the year-on-year sequence data in each preset observation period;

Step 3: Combine different filter sequences corresponding to the same preset observation period to obtain a combined sequence. There are two merging methods used, preferably the sum algorithm is used to merge the filter sequences, and there are three merging methods, which are defined as steps 3.1. Steps 3.2 and 3.3, where step 3.1 is the combination of filtering sequences of various assets classified by asset type, step 3.2 is the combination of filtering sequences of various assets classified by investment risk, and step 3.3 is It is to merge the filter sequences of different types of assets in the same category of assets with investment risk classification, and then merge the preliminary merged sequences corresponding to each type of asset;

Step 4: Extrapolate the combined sequence, that is, perform Gaussian filter extrapolation on the combined sequence to obtain an extrapolated combined sequence;

Step 5: Input the year-on-year sequence data and the combined sequence into the linear regression model to obtain the linear regression coefficients. Step 5 includes two different steps, namely Step 5.1 and Step 5.2. Step 5.1 uses the synchronized sequence data corresponding to the asset as the dependent variable. And the combined sequence is used as the independent variable to perform linear regression with intercept term to obtain the linear regression coefficient, and step 5.2 is to use the difference of the synchronized sequence data corresponding to the asset as the dependent variable and the combined sequence as the independent variable to perform linear regression with intercept term , Get the linear regression coefficient;

Step 6: Extrapolate the combined sequence as the independent variable and linear regression coefficient as the coefficient, perform fitting prediction on the asset synchronization sequence data to obtain the predicted value, and obtain the current time, that is, the fitted value of the current period, through the predicted value and fitting Make the difference between the value and adjust the lag period to get the predicted rise and fall;

Step 7: Sort all kinds of assets according to the magnitude of the predicted rise and fall, so as to determine the investment weights corresponding to all kinds of assets;

Step 8: Generate asset investment advice information based on the investment weight of various assets.

Since step 1 includes two processing situations, step 3 includes three processing situations, and step 5 includes two processing situations, the different processing situations of these three steps are sorted and combined, and the investment advice information is the most accurate after actual measurement. Two strategies,

Among them, strategy one is: Step 1.1+Step 2+Step 3.1+Step 4+Step 5.1+Step 6+Step 7+Step 8;

The second strategy is: Step 1.2+Step 2+Step 3.3+Step 4+Step 5.1+Step 6+Step 7+Step 8.

The present application also provides a device for generating asset investment advice information. The device for generating asset investment advice information includes: a memory, a processor, and asset investment advice information stored on the memory and running on the processor The generating program of the asset investment recommendation information is executed by the processor to implement the steps of the method for generating asset investment recommendation information as described in the above embodiment.

The present application also provides a readable storage medium on which a program for generating asset investment advice information is stored. When the program for generating asset investment advice information is executed by a processor, the asset as described in the above embodiment is implemented. Steps of the method of generating investment advice information.

It should be noted that in this article, the terms "include", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements not only includes those elements, It also includes other elements that are not explicitly listed, or elements inherent to the process, method, article, or system. If there are no more restrictions, the element defined by the sentence "including a..." does not exclude the existence of other identical elements in the process, method, article, or system that includes the element.

The serial numbers of the foregoing embodiments of the present application are for description only, and do not represent the superiority or inferiority of the embodiments.

Through the description of the above embodiments, those skilled in the art can clearly understand that the method of the above embodiments can be implemented by means of software plus the necessary general hardware platform. Of course, it can also be implemented by hardware, but in many cases the former is better.的实施方式。 Based on this understanding, the technical solution of this application essentially or the part that contributes to the existing technology can be embodied in the form of a software product, and the computer software product is stored in a storage medium (such as ROM/RAM) as described above. , Magnetic disks, optical disks), including several instructions to make a terminal device (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) execute the methods described in the various embodiments of the present application.

The above are only the preferred embodiments of this application, and do not limit the scope of this application. Any equivalent structure or equivalent process transformation made using the content of the description and drawings of this application, or directly or indirectly used in other related technical fields , The same reason is included in the scope of patent protection of this application.

Claims (9)

1. A method for generating asset investment advice information, wherein the method for generating asset investment advice information includes the following steps:

   Obtain various types of assets to determine the year-on-year sequence data corresponding to the various types of assets, where the classification methods of the various types of assets include asset type classification and investment risk classification;

   Determine the predicted increase and decrease corresponding to each type of asset according to each of the said year-on-year sequence data;

   Determine the investment weights of the various types of assets according to each of the predicted fluctuations.
2. 4. The method for generating asset investment advice information according to claim 1, wherein the step of determining the predicted fluctuation range corresponding to each type of the asset according to each of the year-on-year sequence data comprises:

   Determine the filter sequence corresponding to each type of asset in each preset observation period according to the year-on-year sequence data;

   Combine the filter sequences corresponding to the various types of assets to obtain the combined sequence of each of the preset observation periods, and determine the output length of the combined sequence, so as to modify the output length by the preset length to obtain Extrapolate the combined sequence;

   Input each of the year-on-year sequence data and the combined sequence into a linear regression model to obtain linear regression coefficients;

   According to the extrapolated combined sequence and the linear regression coefficient, a fitting prediction of the same-year sequence data is performed on each type of asset to determine the predicted increase or decrease corresponding to the various types of assets.
3. 3. The method for generating asset investment advice information according to claim 2, wherein the step of combining the filter sequences corresponding to the various types of assets to obtain the combined sequence of each of the preset observation periods comprises:

   Determine investment risk parameters corresponding to various types of assets, and classify the various types of assets according to the investment risk parameters to obtain various sets, and the investment risk parameters corresponding to the various types of assets in the set belong to the same numerical range;

   Merging the filter sequences corresponding to the various types of assets in the collection to obtain an intermediate merging sequence of each of the collections in each of the preset observation periods;

   Merging each of the intermediate merged sequences to obtain a merged sequence corresponding to each of the preset observation periods.
4. 3. The method for generating asset investment recommendation information according to claim 2, wherein the step of inputting each of the year-on-year sequence data and the combined sequence into a linear regression model to obtain linear regression coefficients comprises:

   Each of the year-on-year sequence data is used as a dependent variable, and the combined sequence is used as an independent variable to input into a linear regression model to obtain a linear regression coefficient.
5. 3. The method for generating asset investment advice information according to claim 2, wherein the step of determining the investment weights corresponding to the various types of assets according to each of the predicted fluctuations comprises:

   Sort each of the predicted fluctuations in descending order;

   Determine the investment weight of the asset corresponding to each of the predicted fluctuations according to each of the predicted fluctuations in the order, wherein, the greater the predicted fluctuation, the greater the investment weight of the asset corresponding to the predicted fluctuation Big.
6. The method for generating asset investment recommendation information according to claim 2, wherein the fitting prediction of the year-on-year sequence data is performed on each type of asset according to the extrapolated combined sequence and the linear regression coefficient to determine each The steps of predicting the corresponding rise and fall of the asset class include:

   According to the extrapolated combined sequence and the linear regression coefficient, perform a fitting prediction of the same-year sequence data for each type of asset to determine the predicted value corresponding to the various types of assets;

   Determine the fitting values corresponding to the various types of assets at the current time;

   The predicted value and the fitted value corresponding to each type of asset are differentiated to obtain the predicted increase or decrease corresponding to each type of asset.
7. The method for generating asset investment advice information according to any one of claims 2-5, wherein the step of determining the filter sequence corresponding to each type of asset in each preset observation period according to the year-on-year sequence data comprises:

   Performing zero padding on the year-on-year sequence data, and performing Fourier transform on the year-on-year sequence data after zero padding to obtain corresponding frequency domain data;

   Determine a set of filter coefficients according to each of the preset observation periods, and obtain an intermediate sequence according to the filter coefficients and the frequency domain data;

   Perform an inverse Fourier transform on the intermediate sequence to obtain a filter sequence of the asset in each of the preset observation periods.
8. A device for generating asset investment advice information, wherein the device for generating asset investment advice information includes: a memory, a processor, and generation of asset investment advice information stored on the memory and running on the processor A program, when the asset investment recommendation information generation program is executed by the processor, the steps of the asset investment recommendation information generation method according to any one of claims 1 to 7 are realized.
9. A readable storage medium, wherein a program for generating asset investment advice information is stored on the readable storage medium, and when the program for generating asset investment advice information is executed by a processor, the implementation is as in any one of claims 1 to 7 The steps of the method for generating asset investment advice information described in the item.

Images