CN115982229A

CN115982229A - Security processing method and device, electronic equipment and storage medium

Info

Publication number: CN115982229A
Application number: CN202211574904.6A
Authority: CN
Inventors: 李剑戈; 殷宪晨; 马金龙; 赵瑞; 汪春晓; 曹震
Original assignee: China Securities Co Ltd
Current assignee: China Securities Co Ltd
Priority date: 2022-12-08
Filing date: 2022-12-08
Publication date: 2023-04-18

Abstract

The embodiment of the invention provides a security processing method, a security processing device, electronic equipment and a storage medium, wherein the method comprises the following steps: calculating an auxiliary evaluation value related to the correlation between every two securities in the securities by using the specified description data of the securities; constructing a relational graph by using the auxiliary evaluation values about the correlation of every two securities; identifying each target vertex pair meeting a preset screening condition from the relationship graph to obtain a security pair represented by each target vertex pair; calculating the price correlation coefficient of the securities in each security pair in different historical time periods, and determining the change trend of the price correlation coefficient of each security pair in the time dimension by using the calculated price correlation coefficient; selecting a target security pair with an increased price correlation coefficient represented by a corresponding change trend from a plurality of security pairs; at least one group of securities having a correlation is determined based on the selected target pair of securities. Through the scheme, securities with relevance can be screened out.

Description

Security processing method and device, electronic equipment and storage medium

Technical Field

The present invention relates to the field of data processing technologies, and in particular, to a method and an apparatus for processing securities, an electronic device, and a storage medium.

Background

For any two securities, analyzing whether the securities have relevance has important influence on risk management, investment decision and data analysis of the securities, for example, when buying a combination of securities, the securities with relevance are generally not recommended to be bought at the same time, thereby achieving the purpose of avoiding investment risk. Therefore, there is a need to screen out securities having a correlation from a plurality of securities.

For any two securities, there is typically some association in some description dimension, such as: two securities co-occur in the news, or the institutional flow of funds for the two securities is the same, and so on.

Therefore, based on the above characteristics, how to select related securities from a plurality of securities is a problem to be urgently solved.

Disclosure of Invention

The embodiment of the invention aims to provide a method, a device, an electronic device and a storage medium for processing securities so as to realize the screening of securities with correlation from a plurality of securities. The specific technical scheme is as follows:

in a first aspect, an embodiment of the present application provides a security processing method, including:

calculating an auxiliary evaluation value related to the correlation between every two securities in the securities by using the specified description data of the securities; wherein the specified description data is data capable of characterizing associations that exist between securities;

constructing a relational graph by using the auxiliary evaluation values about the correlation of every two securities; wherein each vertex in the relational graph is used for representing one security in the plurality of securities, and the weight of an edge between any two vertexes is an auxiliary evaluation value related to the correlation between the securities represented by the any two vertexes;

identifying each target vertex pair meeting a preset screening condition from the relational graph to obtain the security pair represented by each target vertex pair; the preset screening condition is that the weight of the edges is greater than a preset threshold value, or after the weight of each edge is sorted from big to small, the sorting order of the edges is smaller than the preset order threshold value;

calculating the price correlation coefficient of the securities in each security pair in different historical time periods, and determining the change trend of the price correlation coefficient of the security pair in the time dimension by using the calculated price correlation coefficient to obtain the corresponding change trend of the security pair;

selecting a target security pair with a corresponding change trend representing price correlation coefficient increased from a plurality of security pairs;

determining at least one group of securities having a correlation based on the selected target pair of securities; wherein each group of securities having a correlation is two securities in a selected one of the target pairs of securities.

Optionally, the specified description data has a plurality of types;

the method for calculating the auxiliary evaluation value about the correlation between every two securities in the plurality of securities by using the specified description data of the plurality of securities comprises the following steps:

for each kind of designated description data of a plurality of securities, calculating evaluation values related to the correlation between every two securities in the plurality of securities by using the kind of designated description data to obtain initial evaluation values corresponding to every two securities of the kind of designated description data;

and carrying out weighted calculation on the initial evaluation value corresponding to each two securities by utilizing the weight corresponding to various specified description data to obtain an auxiliary evaluation value related to the correlation of each two securities.

Optionally, the specified description data has a plurality of types;

calculating an auxiliary evaluation value about correlation between each two securities of the plurality of securities aiming at the specified description data of the target kind by utilizing the specified description data of the target kind aiming at the specified description data of each target kind; wherein the specified description data of each target category comprises at least one kind of specified description data;

the method for constructing the relationship graph by using the auxiliary evaluation value about the correlation of every two securities comprises the following steps:

for the specified description data of each target category, constructing a relational graph by using the auxiliary evaluation value about the correlation between each two securities in the plurality of securities aiming at the specified description data of the target category;

said determining at least one group of securities having a correlation based on the selected target pair of securities comprises:

and searching for repeated target security pairs from the selected target security pairs to obtain at least one group of related securities.

Optionally, the designated description data of the plurality of securities is data of a target time period; the at least one group of securities having a correlation is a security having a correlation belonging to the target time period;

the method further comprises the following steps:

acquiring securities with correlation in each group belonging to a historical time period to obtain each reference securities group; wherein the historical time period is a time period before the target time period;

determining the reference securities groups with the same rise and fall by using the securities prices of the securities in each reference securities group in the target time period, and occupying the proportion of each reference securities group to obtain a target proportion;

determining the target proportion, and determining the estimated probability of the same rise and fall of at least one group of related securities in the target time period in a future time period; wherein the future time period is a time period after the target time period.

Optionally, the constructing a relationship graph by using the secondary evaluation values about the relevance of each two securities further includes:

and displaying the constructed relation diagram according to a preset relation diagram display mode.

In a second aspect, an embodiment of the present application provides a security processing apparatus, comprising:

the calculation module is used for calculating an auxiliary evaluation value related to the correlation between every two securities in the securities by using the specified description data of the securities; wherein the specified description data is data capable of characterizing associations existing between securities;

the building module is used for building a relational graph by using the auxiliary evaluation values of every two securities about the correlation; wherein each vertex in the relational graph is used for representing one security in the plurality of securities, and the weight of an edge between any two vertexes is an auxiliary evaluation value related to the correlation between the securities represented by the any two vertexes;

the identification module is used for identifying each target vertex pair meeting the preset screening condition from the relationship graph to obtain the security pair represented by each target vertex pair; the preset screening condition is that the weight of the edges is greater than a preset threshold value, or after the weight of each edge is sorted from big to small, the sorting order of the edges is smaller than the preset order threshold value;

the first determining module is used for calculating the price correlation coefficient of the securities in the securities pair in different historical time periods aiming at each security pair, and determining the change trend of the price correlation coefficient of the security pair in the time dimension by using the calculated price correlation coefficient to obtain the corresponding change trend of the security pair;

the selecting module is used for selecting a target security pair with an increased price correlation coefficient represented by a corresponding change trend from a plurality of security pairs;

a second determination module for determining at least one group of securities having a correlation based on the selected target pair of securities; wherein each group of securities having a correlation is two securities in a selected one of the target pairs of securities.

Optionally, the specified description data has a plurality of types;

the calculation module comprises:

a first calculation unit, configured to calculate, for each type of designated description data of a plurality of securities, an evaluation value regarding correlation between each two securities in the plurality of securities by using the type of designated description data, and obtain an initial evaluation value corresponding to each two securities for the type of designated description data;

and the second calculation unit is used for performing weighted calculation on the initial evaluation values corresponding to each two securities by using the weights corresponding to various kinds of specified description data to obtain auxiliary evaluation values of each two securities about the correlation.

Optionally, the specified description data has a plurality of types;

the calculation module comprises:

a third calculation unit configured to calculate, for the specified description data of each target category, an auxiliary evaluation value regarding a correlation between each two securities of the plurality of securities, for the specified description data of the target category, using the specified description data of the target category; wherein the specified description data of each target category comprises at least one kind of specified description data;

the building module comprises:

the construction unit is used for constructing a relational graph by utilizing the auxiliary evaluation value about the correlation between each two securities in the securities aiming at the specified description data of each target class;

the second determining module includes:

and the searching unit is used for searching the repeated target security pairs from the selected target security pairs to obtain at least one group of related securities.

In a third aspect, an embodiment of the present application provides an electronic device, including a processor, a communication interface, a memory, and a communication bus, where the processor and the communication interface complete communication between the memory and the processor through the communication bus;

a memory for storing a computer program;

and the processor is used for realizing any security processing method step when executing the program stored in the memory.

In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, and when being executed by a processor, the computer program implements any of the security processing method steps described above.

The embodiment of the invention has the following beneficial effects:

according to the security processing method provided by the embodiment of the invention, at least one group of related securities can be selected from a plurality of securities by combining the specified description data and the price correlation coefficient of the securities in the security pair. Therefore, through the scheme, the purpose of screening out the securities with correlation from a plurality of securities can be achieved.

Of course, not all of the advantages described above need to be achieved at the same time in the practice of any one product or method of the invention.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other embodiments can be obtained by referring to these drawings.

FIG. 1 is a flow chart of a security processing method provided by an embodiment of the present invention;

FIG. 2 is another flow chart of a security processing method according to an embodiment of the present invention;

FIG. 3 is another flow chart of a security processing method according to an embodiment of the present invention;

FIG. 4 is another flow chart of a security processing method according to an embodiment of the present invention;

FIG. 5 is a schematic view of a security processing apparatus according to an embodiment of the present invention;

fig. 6 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention are within the scope of the present invention.

In order to solve the problem of how to screen out securities with relevance from a plurality of securities, embodiments of the present invention provide a method, an apparatus, an electronic device, and a storage medium for processing securities.

The security processing method provided by the embodiment of the invention can comprise the following steps:

calculating an auxiliary evaluation value about correlation between every two securities in a plurality of securities by using specified description data of the securities; wherein the specified description data is data capable of characterizing associations that exist between securities;

In the scheme, at least one group of securities with correlation can be selected from a plurality of securities by combining the specified description data and the price correlation coefficient of the securities in the securities pair. Therefore, through the scheme, the purpose of screening out the securities with correlation from a plurality of securities can be achieved.

A security processing method according to an embodiment of the present invention will be described with reference to the accompanying drawings.

As shown in fig. 1, the security processing method may include the steps of:

s101, calculating an auxiliary evaluation value about correlation between each two securities in a plurality of securities by using specified description data of the securities; wherein the specified description data is data capable of characterizing the existing associations between securities;

illustratively, the specified descriptive data may include, but is not limited to, securities news data, fund flow data, financing and financing data, volume data, or financial data, among others. The securities news data, the fund flow data, the financing and financing coupon data, the volume price data and the financial data belong to different kinds of specified description data, and each kind of specified description data is data capable of representing the association existing in one description dimension among the securities. It is understood that it is reasonable to use only one kind of specific description data or a plurality of kinds of specific description data when calculating the auxiliary evaluation value regarding the correlation between each two securities of the plurality of securities.

Wherein, for different kinds of specified description data, different calculation modes can be adopted to calculate the auxiliary evaluation value about the correlation between every two securities. For example, when the description data is specified to relate only to the news data of securities, the secondary evaluation value regarding the correlation between each two securities may be the number of news that collectively contain the two securities, for example: for the securities a and B, the news numbers that collectively contain the securities a and B will be used as an auxiliary evaluation value for correlation between the securities a and B. Illustratively, when the description data only relates to the fund flow data, the secondary evaluation value about the correlation between every two securities is a predetermined statistical time interval, and the proportion of the days of the common flow is, for example, the net inflow data of funds of two securities A and B in the past n days is counted, if the current day is positive, the current day is 1, the current day is negative, the current day is-1, each security can obtain an n-dimensional vector, and the number ratio of the same number between the same dimensions of the two security vectors is counted, and the secondary evaluation value about the correlation between the securities A and B is counted. Illustratively, when the description data is only related to financing instrument data, the auxiliary evaluation value of the relevance between every two instruments is the proportion of days with the same change direction of the financing instrument data in the statistical time interval, for example, two instruments A and B financing instrument data of the past n days are counted and subtracted from the previous day, if the current day is positive, the current day is 1, the current day is negative, the current day is-1, otherwise, the current day is 0, each instrument can obtain an n-dimensional vector, and the number ratio of the same number between the same dimensions of the two instrument vectors is counted and the auxiliary evaluation value of the two instruments A and B is counted. Illustratively, when the description data relates to data with time sequence properties such as measurement price, finance and the like, the auxiliary evaluation value related to the correlation between every two securities is a similarity value between two security time sequence data composition vectors within a statistic time interval, the similarity value is marked as the auxiliary evaluation value of the two securities, and the similarity calculation mode comprises a multidimensional vector measurement method not limited to a Pearson correlation coefficient and cosine similarity. In the present invention, the calculation method of the auxiliary evaluation value is not limited, and any calculation method of the auxiliary evaluation value may be applied to the present invention.

It should be noted that, if the type of the specified description data is one, the auxiliary evaluation value about the correlation between each two securities in the plurality of securities can be determined by using a corresponding calculation method for the specified description data, for example, the method exemplarily given above. If the types of the appointed description data in the scheme are various, in one implementation mode, the auxiliary evaluation value about the correlation between every two securities in the multiple securities can be calculated by combining various appointed description data; in another embodiment, for each specific description data, an initial evaluation value between two securities of the plurality of securities can also be determined using the specific description data, so that for each two securities there are a plurality of auxiliary evaluation values relating to relevance. For clarity of layout and clarity of the scheme, the following describes specific implementations when the description data is specified in various types in combination with other embodiments.

S102, constructing a relational graph by using the auxiliary evaluation values of every two securities about the correlation; wherein each vertex in the relational graph is used for representing one security in the plurality of securities, and the weight of an edge between any two vertexes is an auxiliary evaluation value related to the correlation between the securities represented by the any two vertexes;

the relationship graph can represent the relationship between the vertexes and the vertexes, and the specific relationship is represented by the weight of the edge between the vertexes, the vertexes in the constructed relationship graph are used for representing the securities in the plurality of securities, the weight of the edge between any two vertexes is the auxiliary evaluation value related to the correlation between the securities represented by any two vertexes, therefore, the constructed relationship graph can be used for representing the relationship of the securities with the specified description data corresponding to the auxiliary evaluation value. Illustratively, there is a relationship map in which the secondary evaluation value is calculated from the security news data, and this relationship map may represent the relationship of association between the securities in the map with respect to the security news data.

Optionally, the step of constructing a relationship graph by using the secondary evaluation values about the relevance of each two securities further includes:

In one implementation, the data of the relational graph can be acquired after the relational graph is constructed and stored in the database; and when the relation graph needs to be displayed, acquiring the data of the relation graph from the database and displaying the relation graph according to a set display mode. The set display mode may be a shape in which vertices in the relational graph are connected into a polygon, and the display mode is not limited herein.

The relational graph constructed by the specified description data can lead investors to know the relation of the specified description data among a plurality of securities and provide reference for risk management, investment decision and data analysis of the investors.

S103, identifying each target vertex pair meeting a preset screening condition from the relational graph to obtain a security pair represented by each target vertex pair; the preset screening condition is that the weight of the edges is greater than a preset threshold value, or after the weight of each edge is sorted from big to small, the sorting order of the edges is smaller than the preset order threshold value;

the condition meeting the predetermined screening condition is a condition which needs to be met by screening securities with correlation, and the other condition is that the price correlation coefficient of the securities meets the condition, which will be elaborated in the subsequent embodiments and will not be described in detail herein; the relationship diagram only represents the relationship between the securities represented by the vertexes with respect to the specified description data, so that securities having relevance are first screened out of the pairs of securities represented by the target vertexes meeting the predetermined screening condition. The condition of meeting the preset screening condition means that the incidence relation of the securities relative to the specified description data meets the preset condition; in the relationship graph, the association relationship is characterized by the weight of the edge, so the predetermined filtering condition may be that the weight of the edge is greater than a predetermined threshold, or after the weights of the edges are sorted from large to small, the sorting order of the edges is smaller than the predetermined order threshold, and the predetermined filtering condition may be adaptively adjusted for different types of securities and different types of specified description data, and the specific form of the predetermined filtering condition is not limited by the present scheme.

For example, there are 50 vertex pairs in a relation graph related to security news data, the predetermined filtering condition is that the weight of the edge is greater than the predetermined threshold 5, and the weight of 10 vertex pairs in the relation graph is greater than 5, then the 10 vertex pairs meet the predetermined filtering condition; in another implementation, there is a relationship diagram about the fund flow direction data, and there are 100 target vertex pairs in the diagram, and the predetermined screening condition is that after the weights of the respective edges are sorted from large to small, the sorting order of the edges is smaller than the predetermined order threshold 11, then the weights of the edges of the 100 target vertex pairs are sorted from large to small, the 10 target vertex pairs with the largest edge weight are sorted in 1-10, and the sorting order is smaller than the threshold 11, then the 10 target vertex pairs sorted in 1-10 meet the predetermined screening condition.

S104, calculating the price correlation coefficient of the securities in each security pair in different historical time periods, and determining the change trend of the price correlation coefficient of the security pair in the time dimension by using the calculated price correlation coefficient to obtain the corresponding change trend of the security pair;

the correlation coefficient is a statistical index for reflecting the degree of closeness of the correlation between the variables, and the price correlation coefficient of the securities is a statistical index for reflecting the degree of closeness of the correlation between the prices of the securities. The price correlation coefficient between two securities is calculated according to price data of the two securities in a time period, and the price data of the securities comprises data such as transaction amount, volume of transaction, closing price, rate of return and the like; the calculation method of the price correlation coefficient may use a pearson correlation coefficient, a cosine similarity, and other manners, exemplarily, the stock price data of the stock pair is the closing price, and then the calculation method of the cosine similarity may be used to calculate the similarity between vectors formed by the closing price data of the stocks in the stock pair in a certain historical time period, so as to obtain the price correlation coefficient of the stock pair in the historical time period. The method for calculating the price correlation coefficient is not limited, and any method for calculating the price correlation coefficient can be applied to the method.

Wherein the price correlation coefficient of each security pair with respect to different historical time periods can be calculated using the price data of the different historical time periods, and the trend of the change of the price correlation coefficient of the relatively new time period to the price correlation coefficient of the relatively old time period of each security pair can be determined using the price correlation coefficients. The variation trend of the price correlation coefficient is in positive correlation with the variation trend of the degree of closeness of the correlation of the price.

For example, there are two time periods, one is day 17 to 21 of 10 months, and one is day 10 to 14 of 10 months, and each is day 5 of 10 months, 10 days to 14 days of 10 months, for the 10 security pairs obtained through the above steps, the price correlation coefficient of the security pair in each time period is respectively calculated, and the variation trend of the price correlation coefficient of the relatively new time period of each security pair with respect to the price correlation coefficient of the relatively old time period is determined, for example, the price correlation coefficient of the relatively new time period of each security pair is subtracted from the price correlation coefficient of the relatively old time period, if positive, the variation trend of the security pair is increased, and the determination method of the variation trend of the price correlation coefficient is not limited thereto, and any method for determining the variation trend of the price correlation coefficient may be applied to the present invention.

S105, selecting a target security pair with an increased price correlation coefficient represented by a corresponding change trend from the plurality of security pairs;

the price correlation coefficient variation trend shows that the increasing of the degree of affinity of the correlation between the securities shows that the securities have the correlation, so that the selection of the target security pair with the corresponding variation trend representing the increasing of the price correlation coefficient, namely the selection of the target security pair representing the degree of affinity of the price correlation, at the moment, the screening of the auxiliary evaluation value of the securities and the screening of the price correlation coefficient can obtain the target security pair, and the selected target security pair can be considered to have the correlation.

For example, the above embodiment obtains the trend of change of the price correlation coefficients of 10 security pairs related to news data, and uses the trend of change of the price correlation coefficients to select a target security pair with a trend of change representing an increase in the price correlation coefficients, and in the above 10 security pairs, there are 3 security pairs with an increase in the trend of change of the price correlation coefficients, and then these three security pairs are the selected target security pairs.

S106, determining at least one group of securities with correlation based on the selected target security pair; wherein each group of securities having a correlation is two securities in a selected one of the target pairs of securities.

In one implementation, if a relationship graph is constructed for one or more specified description data, the selected target security pairs can be directly used as a group of related securities. That is, two securities in each target pair selected in the above step have correlation, and in the above embodiment, three target pairs of securities with respect to news data, i.e., a-D, a-E, and N-F, are selected, and it can be determined that at least securities a and D have correlation, securities a and E have correlation, and securities N and F have correlation.

Optionally, the specified description data of the securities is data of a target time period; the at least one group of securities having a correlation is a security having a correlation belonging to the target time period;

the method further comprises steps a-C:

step A, obtaining securities with correlation in each group belonging to a historical time period to obtain each reference securities group; wherein the historical time period is a time period before the target time period;

the method of determining securities with correlations belonging to historical time periods described herein is similar to that described above, except that the determined securities belong to different time periods. The cycle duration of the target time period and the historical time period may be set according to actual conditions, for example: the cycle duration may be one month, one week, one day, etc., which is reasonable.

Step B, determining the reference securities groups with the same rise and fall by using the securities prices of the securities in each reference securities group in the target time period, and occupying the proportion of each reference securities group to obtain a target proportion;

the total price change trend of two related securities in the reference security group in the target time period is the same, and then the reference security group can be considered to rise and fall at the same time, or, in the target time period, the proportion of the number of days with the same price change trend among the related securities in the reference security group in the total number of days is larger than the set threshold number of days, and then the reference security group is considered to rise and fall at the same time.

Illustratively, with t as the calculation day, which is the date of calculation using the document processing method of the present application, and 5 days as the time period, the price of A is rising on the day t-1 relative to the day t-5, and the price of B is rising on the day t-1 relative to the day t-5, for the reference document set A-B, in the time period from t-5 to t-1, the reference document set A-B is considered to be rising and falling at the same time; in another implementation, the reference set of securities A-B is considered to be in-process and in-process with the price of A and B rising on a day t-5, the price of A and B falling on a day t-3, the price of A and B rising on a day t-2, the price of A and B falling on a day t-1, the price of A and B rising or falling on a day B and B, respectively, each of the prices of A and B rising or falling days greater than a set three-day threshold.

Step C, determining the target proportion as the estimated probability of the same rise and fall of at least one group of related securities in the target time period in the future time period; wherein the future time period is a time period after the target time period.

For example, the future time period may be t to t +4, the target time period may be t-5 to t-1, and the historical time period may be t-10 to t-6, then the number of pairs of securities having the same rise and fall in the pair of securities having the same rise and fall in the time period from the target time period t-5 to t-1 and the historical time period t-10 to t-6 may be calculated according to the price data of the securities in the two time periods from the target time period t-5 to t-1, and the proportion of pairs of securities having the same rise and fall in the historical time period to the pair of securities having the same fall in the historical time period may be calculated, a target proportion is obtained, and the target proportion is used as the estimated probability that the pair of securities having the same fall in the time period from the target time period t-5 to t-1 in the future time period t-4.

The target proportion is used as the estimated probability of the same rise and fall of the securities with correlation in a future time period and can be used as reference data, so that the investor has reference basis when investing the securities in the reference securities group.

In this embodiment, at least one group of securities having a correlation can be selected from a plurality of securities by combining the specified description data and the price correlation coefficient of the securities in the pair of securities. Therefore, the scheme can realize the purpose of screening out related securities from a plurality of securities.

Alternatively, in another embodiment, as shown in FIG. 2, a method of processing securities may comprise the steps of:

s201, aiming at each kind of designated description data of a plurality of securities, calculating an evaluation value related to the correlation between every two securities in the plurality of securities by using the designated description data to obtain an initial evaluation value corresponding to every two securities aiming at the designated description data;

in this embodiment, the plurality of types of the designated description data are provided, and each type of the designated description data is data capable of characterizing an association existing in one description dimension between securities.

It will be appreciated that there may be more than one specific description data for each security, and may include, for example, but not limited to, security news data, fund flow data, financing instrument data, pricing data or financial data, and the like. The stock news data, the fund flow data, the financing and financing note data, the volume price data and the financial data belong to different kinds of specified description data. It is to be understood that an initial evaluation value regarding the correlation corresponding to each of the specified description data between each two securities of the plurality of securities may be calculated for each of the specified description data. For example, if the securities A and B relate to three specific description data of news co-occurrence, institution fund flow direction and banker fund flow direction, three initial evaluation values related to the correlation of the news co-occurrence, the institution fund flow direction and the banker fund flow direction can be calculated between the securities A and B; the present invention does not limit the number of kinds of the specified description data and the number of securities.

S202, carrying out weighted calculation on the initial evaluation values corresponding to each two securities by utilizing the weights corresponding to various specified description data to obtain the auxiliary evaluation value of each two securities about the correlation.

The initial evaluation values corresponding to each two securities are multiple, the multiple initial evaluation values are weighted and calculated by using weights corresponding to various kinds of specified description data, and auxiliary evaluation values related to correlation of each two securities can be obtained. The weight corresponding to each kind of designated description data may be a weight set according to the importance degree and/or the demand degree of the designated description data, and is not limited herein. The auxiliary evaluation value is an evaluation value of the plurality of specified description data about the relevance, so the scheme can simultaneously analyze the relevance of the plurality of specified description data of every two securities, and the dimensionality of the specified description data when the securities with the relevance are screened is expanded.

S203, constructing a relational graph by using the auxiliary evaluation values of every two securities about the correlation; wherein each vertex in the relational graph is used for representing one security in the plurality of securities, and the weight of an edge between any two vertexes is an auxiliary evaluation value related to the correlation between the securities represented by the any two vertexes;

s204, identifying each target vertex pair meeting the preset screening condition from the relational graph to obtain the security pair represented by each target vertex pair; the preset screening condition is that the weight of the edges is greater than a preset threshold value, or after the weights of all the edges are sorted from large to small, the sorting order of the edges is smaller than a preset order threshold value;

s205, calculating the price correlation coefficient of the securities in the securities pair in different historical time periods aiming at each securities pair, and determining the change trend of the price correlation coefficient of the securities pair in the time dimension by using the calculated price correlation coefficient to obtain the corresponding change trend of the securities pair;

s206, selecting a target security pair with an increased price correlation coefficient represented by a corresponding change trend from the plurality of security pairs;

s207, determining at least one group of securities with correlation based on the selected target security pair; wherein each group of securities having a correlation is two securities in a selected one of the target pairs of securities.

Wherein steps S203-S208 are the same as steps S102-S106 in the above embodiment, and are not described again here.

In this embodiment, at least one group of securities having correlation can be selected from a plurality of securities by combining the specified description data and the price correlation coefficient of the securities in the pair of securities. Therefore, the scheme can realize the purpose of screening out related securities from a plurality of securities.

In addition, the auxiliary evaluation value in the embodiment is a correlation evaluation value for a plurality of different kinds of designated description data, so that securities with correlation can be screened out by using the plurality of different kinds of designated description data, the dimensionality of the designated description data is expanded, and the accuracy of screening out the securities with correlation is improved.

Alternatively, in another embodiment, as shown in FIG. 3, a method of processing securities may comprise the steps of:

s301, aiming at the specified description data of each target type, calculating an auxiliary evaluation value about the correlation between each two securities in the plurality of securities aiming at the specified description data of the target type by using the specified description data of the target type; wherein the specified description data of each target category comprises at least one kind of specified description data;

in this embodiment, a plurality of types of the description data are specified, and each type of the description data is data capable of characterizing an association existing in one description dimension between securities.

It is understood that there may be a plurality of specific description data for each security, and for example, the specific description data may include, but is not limited to, security news data, fund flow data, financing instrument data, volume price data, or financial data, etc., and a secondary evaluation value of the correlation between the specific description data of a target category between each two securities may be calculated using the specific description data of the target category between each two securities, wherein the target category refers to a grouping of the specific description data, each specific description data of the target category includes at least one specific description data, the specific description data of a plurality of target categories may exist for the same security pair, and for example, there are three specific description data of security news data, fund flow data, and financing instrument data for the security pairs a-B, and there may also be specific description data of one target category including security news data, fund flow data, and financing instrument data.

If only one kind of the designated description data is included in the designated description data of one target category, the step of calculating the secondary evaluation value regarding the correlation between each two securities in the plurality of securities calculates the secondary evaluation value regarding the correlation for the designated description data by using the designated description data of the plurality of securities in the above embodiment. If the specified description data of one target category comprises a plurality of kinds of specified description data, calculating evaluation values related to the relevance between every two securities in the plurality of securities according to each kind of specified description data of the plurality of securities in the embodiment, and obtaining initial evaluation values corresponding to every two securities of the specified description data; and a step of performing weighted calculation on the initial evaluation value corresponding to each two securities by using the weight corresponding to various specified description data to obtain an auxiliary evaluation value related to the relevance of each two securities, so as to obtain an auxiliary evaluation value related to the relevance of a plurality of specified description data of each two securities.

S302, aiming at the specified description data of each target type, constructing a relational graph by utilizing the auxiliary evaluation value about the correlation between each two securities in the securities aiming at the specified description data of the target type;

wherein, due to the difference of the specified description data of the target category, the relationship graph can represent the relationship between the securities with respect to the different specified description data. Thus, for a same security pair, there may be a relationship graph constructed from one or more of its various specified description data; illustratively, the security pair a-B has a plurality of specified description data of security news data, fund flow data and financing data, and for the security pair a-B, there may be a relationship diagram constructed by the specified description data of one target category including the security news data, the relationship diagram characterizing the relationship of the security pair a-B with respect to the security news data, or there may be a relationship diagram constructed by the specified description data of one target category including the security news data, the fund flow data and the financing data, the relationship diagram characterizing the relationship of the security pair a-B with respect to three dimensions of the security news data, the fund flow data and the financing data.

S303, identifying each target vertex pair meeting the preset screening condition from the relational graph to obtain the security pair represented by each target vertex pair; the preset screening condition is that the weight of the edges is greater than a preset threshold value, or after the weights of all the edges are sorted from large to small, the sorting order of the edges is smaller than a preset order threshold value;

s304, calculating the price correlation coefficient of the securities in the securities pairs in different historical time periods aiming at each security pair, and determining the change trend of the price correlation coefficient of the securities pair in the time dimension by using the calculated price correlation coefficient to obtain the corresponding change trend of the securities pair;

s305, selecting a target security pair with an increased price correlation coefficient represented by a corresponding change trend from the plurality of security pairs;

s306, searching repeated target security pairs from the selected target security pairs to obtain at least one group of related securities.

In one implementation, the designated description data corresponding to the relation graph a is securities news data, the selected target securities pair is A-B, D-C and E-F, the designated description data corresponding to the relation graph B is capital flow data, the selected target securities pair is A-B, D-G and E-N, and then repeated securities pair is selected as A-B to obtain a securities pair A-B with correlation; in another implementation mode, the designated description data corresponding to the relation graph a is securities news data, the selected target securities pairs are A-B, D-C and E-F, the designated description data corresponding to the graph C is securities news data, capital flow direction data and financing securities data, the selected target securities pairs are A-C, D-G and N-F, and repeated securities pairs are selected as D-C to obtain related securities pairs D-C; the target security pairs in one relational graph have correlation with the specified description data of the relational graph, and the repeated security pairs in a plurality of relational graphs indicate that the security pairs have correlation with the specified description data corresponding to the plurality of relational graphs, so that the accuracy of screening the security pairs with correlation can be improved by searching the repeated target security pairs from the selected target security pairs. The scheme can also perform correlation analysis on different security relationship graphs to find out security pairs with relevance.

Wherein steps S303 to S305 are the same as steps S103 to S105 in the above embodiment, and are not described again here.

In addition, in the embodiment, the target security pairs selected from different relationship graphs are used for selecting the security pairs which repeatedly appear as the security pairs with the relevance, so that the accuracy of screening the securities with the relevance is improved.

In order to better understand the scheme, the certificate processing method is explained by a specific embodiment; FIG. 4 is a flow chart of a security processing method;

s401, acquiring historical data;

the historical data comprises data required by mapping and securities price data, and the mapping data comprises, but is not limited to, securities news data, fund flow data (fund such as institution, large family, middle family, scattered family, north direction and the like), financing and financing data, volume price data, financial data and the like. Securities price data includes, but is not limited to, transaction amount, volume, closing price, rate of return, etc. The mapping data is the specific description data in the above embodiment, and the securities price data is the data required for calculating the price correlation coefficient in the above embodiment, such as the data of transaction amount, volume of transaction, closing price, rate of return, and the like. The period duration of the historical data may be set according to actual conditions, for example: the period duration may be, for example, a week analysis frequency, which includes 5 transaction days, and a month frequency analysis frequency, which includes 21 transaction days, and the like, which are all reasonable, and the selection of the period duration is not limited. In one implementation, taking t as a calculation day, which is a date calculated by using the security processing method of the present application, and a time period may be 5 days, a data time interval required for mapping may be t-10 to t-6, and the security price data may be data of two time intervals, i.e., t-10 to t-6 and t-5 to t-1, where t-10 to t-6 are historical time periods, t-5 to t-1 are target time periods, and the historical time periods are time periods before the target time periods.

S402, constructing a security relationship graph;

the step of constructing the relationship diagram is similar to the step of constructing the relationship diagram by using the auxiliary evaluation value about the correlation of each two securities, namely, the step of calculating the auxiliary evaluation value by using the data of the historical time period to construct the relationship diagram, which is not described in detail herein.

In one implementation, the calculation date may be t, the calculation date is the date of calculation by the security processing method of the present application, the time period is 5 days, and the historical time period may be t-10 to t-6, i.e., 5 days of data are mapped in the manner of the above-described map. The relationship graph constructed in this step is a historical target relationship graph.

S403, filtering graph vertexes and edges;

the graph vertex and edge filtering step is similar to the step of identifying target vertex pairs meeting the predetermined filtering condition from the relationship graph to obtain the security pairs represented by each target vertex pair, and the step can map the selected target vertex pairs to obtain the relationship graph related to the relevance of the assigned description data.

S404, storing and visually displaying a graph database;

the data of the target relational graph is stored in the graph database and can be displayed in a set mode, and illustratively, the relational graph can be visually displayed by taking securities in the target relational graph as vertexes of a polygon.

S405, calculating and analyzing the securities price correlation in the graph;

by the securities price data, the securities price correlation in the relational graph can be calculated, illustratively, the calculation date can be t, the calculation date is the date of calculation by using the securities processing method, 5 days can be taken as a time period, and the calculation time interval of securities to the price correlation data is data of two time intervals of t-10 to t-6 and t-5 to t-1. And finding out the securities pair with enhanced price correlation coefficient characterization as the securities pair with correlation according to the securities pair price correlation data of the two time intervals. And calculating the same rise and fall probability of the security pair in the time interval from t-5 to t-1 in the target relational graph according to the price data.

Moreover, the price correlation coefficients of the securities in different time intervals can be visually displayed in a thermodynamic diagram mode, so that investors can more clearly know the change trend of the price correlation coefficients of the securities.

S406, acquiring the latest data;

similar to S401, the difference is that this step only needs to obtain the data required for mapping, and the time interval of the data required for mapping may be a target time period, for example, the target time period may be t-5 to t-1.

S407, constructing a security relationship graph;

similar to S402, the mapping data time interval may be a target time period, for example, the target time period may be t-5 to t-1, that is, for a 5-day time period with the latest date t, the latest target relationship map with the time interval of t-5 to t-1 is finally constructed.

S408, filtering the top points and the edges of the graph;

similar to S403.

S409, storing and visually displaying a graph database;

similar to S404.

S410, analyzing the investment of securities;

in the step, the same-rise and same-fall probabilities of the security pairs in the historical target relational graph in the target time period are firstly determined, for example, if the total variation trends of the price data of the security pairs in the historical target relational graph in the target time period t-5 to t-1 are the same, the security pairs can be considered to be in the same-rise and same-fall, or if the proportion of the number of days with the same variation trend of the price data among the security pairs in the historical target relational graph in the time period t-5 to t-1 to the total number of days is larger than the set threshold number of days, the security pairs are considered to be in the same-rise and same-fall. The ratio of the securities pairs with the same rise and fall in the historical relationship graph can be determined to be the estimated probability of the same rise and fall of at least one group of the securities with the correlation in the latest target relationship graph in the future time period; the future time period is a time period after the target time period.

Illustratively, taking t as a calculation day, which is a date calculated using the security processing method of the present application, 5 days as a time period, and a-B of a security pair in the time period from t-5 to t-1, the price of a is rising on the day t-1 relative to the day t-5, and the price of B is rising on the day t-1 relative to the day t-5, then the reference security group a-B is considered to be rising and falling; in another implementation, the price of A and B rises simultaneously on a day t-5, the price of A and B falls simultaneously on a day t-3, the price of A and B rises simultaneously on a day t-2, the price of A and B falls simultaneously on a day t-1, the price of A and B rises and falls simultaneously on a day t-1, the price of A and B both have a date of rising or falling, four days longer than the set three-day threshold, and the security pair A-B is considered to rise and fall simultaneously. Whether each security pair in the historical target relationship graph is in the same rise and fall can be calculated, the proportion of the security pairs in the historical target relationship graph in the same rise and fall can be calculated, and the proportion is used as the estimated probability of the same rise and fall of each security pair in the latest target relationship graph in the future time period t to t +4 time interval; the scheme can be used as reference data when investors invest in the securities in the latest target relationship graph.

In this embodiment, at least one group of securities having a correlation can be selected from a plurality of securities by combining the specified description data and the price correlation coefficient of the securities in the pair of securities. Therefore, the scheme can realize the purpose of screening out related securities from a plurality of securities. And the ratio of the same-rise and same-fall security pairs in the historical target relationship diagram is used as the same-rise and same-fall probability of the security pairs in the latest target relationship diagram in the future time period, the same-rise and same-fall probability can be used as the reference basis for investing the security in the latest target relationship diagram for the investor to reference, and the security relationship diagram and the price correlation coefficient are visually displayed, so that the investor can more conveniently know the relevant data.

An embodiment of the present invention further provides a security processing apparatus, as shown in fig. 5, the apparatus including:

a calculating module 501, configured to calculate an auxiliary evaluation value related to a correlation between each two securities in a plurality of securities by using specified description data of the securities; wherein the specified description data is data capable of characterizing associations that exist between securities;

a construction module 502, configured to construct a relational graph by using the auxiliary evaluation values about the correlation of every two securities; wherein each vertex in the relational graph is used for representing one security in the plurality of securities, and the weight of an edge between any two vertexes is an auxiliary evaluation value related to the correlation between the securities represented by the any two vertexes;

an identifying module 503, configured to identify, from the relationship diagram, each target vertex pair that meets a predetermined screening condition, to obtain a security pair represented by each target vertex pair; the preset screening condition is that the weight of the edges is greater than a preset threshold value, or after the weight of each edge is sorted from big to small, the sorting order of the edges is smaller than the preset order threshold value;

a first determining module 504, configured to calculate, for each security pair, a price correlation coefficient of the security in the security pair in different historical time periods, and determine, by using the calculated price correlation coefficient, a variation trend of the price correlation coefficient of the security pair in a time dimension, so as to obtain a variation trend corresponding to the security pair;

a selecting module 505, configured to select, from the plurality of security pairs, a target security pair whose corresponding trend represents an increase in the price correlation coefficient;

a second determination module 506 for determining at least one group of securities having a correlation based on the selected target pair of securities; wherein each group of securities having a correlation is two securities in a selected one of the target pairs of securities.

Optionally, the specified description data has a plurality of types;

the calculation module comprises:

Optionally, the specified description data has a plurality of types;

the calculation module comprises:

the building module comprises:

the second determining module includes:

and the searching unit is used for searching repeated target security pairs from the selected target security pairs to obtain at least one group of securities with correlation.

the device further comprises:

the acquisition module is used for acquiring securities with correlation in each group belonging to a historical time period to obtain each reference securities group; wherein the historical time period is a time period before the target time period;

the third determining module is used for determining the reference securities groups with the same rise and fall by using the securities prices of the securities in each reference securities group in the target time period, and the reference securities groups account for the proportion of each reference securities group to obtain a target proportion;

a fourth determining module, configured to determine the target proportion as an estimated probability of a same rise and a same fall of at least one group of securities having a correlation in a future time period within the target time period; wherein the future time period is a time period after the target time period.

Optionally, the apparatus further comprises:

and the display module is used for displaying the constructed relation diagram according to a preset relation diagram display mode.

An embodiment of the present invention further provides an electronic device, as shown in fig. 6, including a processor 601, a communication interface 602, a memory 603, and a communication bus 604, where the processor 601, the communication interface 602, and the memory 603 complete mutual communication through the communication bus 604,

a memory 603 for storing a computer program;

the processor 601 is configured to implement any of the above-described security processing method steps when executing the program stored in the memory 603.

The communication bus mentioned in the electronic device may be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The communication bus may be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one thick line is shown, but this does not mean that there is only one bus or one type of bus.

The communication interface is used for communication between the electronic equipment and other equipment.

The Memory may include a Random Access Memory (RAM) or a Non-Volatile Memory (NVM), such as at least one disk Memory. Alternatively, the memory may be at least one memory device located remotely from the processor.

The Processor may be a general-purpose Processor, including a Central Processing Unit (CPU), a Network Processor (NP), and the like; but also Digital Signal Processors (DSPs), application Specific Integrated Circuits (ASICs), field Programmable Gate Arrays (FPGAs) or other Programmable logic devices, discrete Gate or transistor logic devices, discrete hardware components.

In a further embodiment of the present invention, a computer-readable storage medium is also provided, in which a computer program is stored, which computer program, when being executed by a processor, carries out the steps of any of the above-mentioned security processing methods.

In a further embodiment provided by the present invention, there is also provided a computer program product containing instructions which, when run on a computer, cause the computer to perform the steps of any of the security processing methods of the embodiments described above.

In the above embodiments, the implementation may be wholly or partially realized by software, hardware, firmware, or any combination thereof. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When loaded and executed on a computer, cause the processes or functions described in accordance with the embodiments of the invention to occur, in whole or in part. The computer may be a general purpose computer, a special purpose computer, a network of computers, or other programmable device. The computer instructions may be stored in a computer readable storage medium or transmitted from one computer readable storage medium to another computer readable storage medium, for example, the computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital Subscriber Line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device, such as a server, a data center, etc., that incorporates one or more of the available media. The usable medium may be a magnetic medium (e.g., floppy Disk, hard Disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid State Disk (SSD)), among others.

It should be noted that, in this document, relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrases "comprising a," "...," or "comprising" does not exclude the presence of additional like elements in a process, method, article, or apparatus that comprises the element.

All the embodiments in the present specification are described in a related manner, and the same and similar parts among the embodiments may be referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for embodiments of apparatuses, devices, storage media, etc., since they are substantially similar to the method embodiments, the description is relatively simple, and reference may be made to the partial description of the method embodiments for relevant points.

The above description is only for the preferred embodiment of the present invention, and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention shall fall within the protection scope of the present invention.

Claims

1. A method of processing securities, characterized in that it comprises:

selecting a target security pair with an increased price correlation coefficient represented by a corresponding change trend from a plurality of security pairs;

2. The method according to claim 1, wherein the specified description data is of a plurality of kinds;

and performing weighted calculation on the initial evaluation value corresponding to each two securities by using the weight corresponding to various specified description data to obtain an auxiliary evaluation value about the correlation of each two securities.

3. The method according to claim 1, wherein the specified description data has a plurality of kinds;

aiming at the specified description data of each target type, constructing a relational graph by utilizing the auxiliary evaluation value about the correlation between each two securities in the plurality of securities aiming at the specified description data of the target type;

4. A method according to any of claims 1-3, wherein the specified description data of the plurality of securities is data of a target time period; the at least one group of securities having a correlation is a security having a correlation belonging to the target time period;

the method further comprises the following steps:

determining the reference security groups with the same rise and fall according to the security prices of the securities in the reference security groups in the target time period, and occupying the proportion of the reference security groups to obtain a target proportion;

determining the target proportion, namely the estimated probability of the same rise and fall of at least one group of securities with correlation in the target time period in the future time period; wherein the future time period is a time period after the target time period.

5. The method according to claim 1, wherein said constructing a relationship graph using the secondary evaluation values about the relevance of each two securities further comprises:

6. A security processing apparatus, characterized in that the apparatus comprises:

the identification module is used for identifying all target vertex pairs meeting the preset screening condition from the relational graph to obtain the security pairs represented by each target vertex pair; the preset screening condition is that the weight of the edges is greater than a preset threshold value, or after the weight of each edge is sorted from big to small, the sorting order of the edges is smaller than the preset order threshold value;

the first determination module is used for calculating the price correlation coefficient of the securities in the securities pairs in different historical time periods aiming at each security pair, and determining the change trend of the price correlation coefficient of the securities pair in the time dimension by using the calculated price correlation coefficient to obtain the corresponding change trend of the securities pair;

a second determination module for determining at least one group of securities having a correlation based on the selected target pair of securities; wherein each set of securities having a correlation is two securities of a selected pair of target securities.

7. The apparatus according to claim 6, wherein the specified description data has a plurality of kinds;

the calculation module comprises:

and the second calculation unit is used for performing weighted calculation on the initial evaluation values corresponding to each two securities by utilizing the weights corresponding to various specified description data to obtain the auxiliary evaluation values of each two securities about the relevance.

8. The apparatus according to claim 6, wherein the specified description data is of a plurality of kinds;

the calculation module comprises:

the building module comprises:

the second determining module includes:

9. The electronic equipment is characterized by comprising a processor, a communication interface, a memory and a communication bus, wherein the processor and the communication interface are used for realizing the communication between the processor and the memory through the communication bus;

a memory for storing a computer program;

a processor for implementing the method steps of any one of claims 1 to 5 when executing a program stored in the memory.

10. A computer-readable storage medium, characterized in that a computer program is stored in the computer-readable storage medium, which computer program, when being executed by a processor, carries out the method steps of any one of the claims 1-5.