CN108492189A - Portfolio Selection selection method, device and equipment - Google Patents
Portfolio Selection selection method, device and equipment Download PDFInfo
- Publication number
- CN108492189A CN108492189A CN201810242799.3A CN201810242799A CN108492189A CN 108492189 A CN108492189 A CN 108492189A CN 201810242799 A CN201810242799 A CN 201810242799A CN 108492189 A CN108492189 A CN 108492189A
- Authority
- CN
- China
- Prior art keywords
- investment combination
- value
- earning rate
- history
- selection
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/06—Asset management; Financial planning or analysis
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Human Resources & Organizations (AREA)
- Entrepreneurship & Innovation (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Technology Law (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The present invention provides a kind of Portfolio Selection selection methods, device and equipment, it is related to the technical field of investment securities, this method is using investment combination as research object, get the history feature value of history investment combination, pass through machine learning model, relationship between research history characteristic value and historical return, predict recent earning rate, again by minimizing difference between last-period forecast earning rate and recent effective yield, machine learning model is optimized, the predicted value of next period earning rate is produced finally by the machine learning model of current characteristic value and optimization, it is selected as desired for user.The selection of Portfolio Selection is carried out as unit of investment combination, optimal Portfolio can be selected from all possible investment combination.
Description
Technical field
The present invention relates to investment securities technical field, more particularly, to a kind of Portfolio Selection selection method, device and
Equipment.
Background technology
It is well known that the investment target in financial market is numerous.With economic and finance development, investment target more they tends to more
Sample.For A share market, a total of more than 2800 stock of A share market at present, then being demonstrate,proved caused by this 2800 stocks
Just there are about as many as 10 33 powers for certificate investment combination quantity.The investment combination of this 10 33 power is referred to as universe investment combination,
English abbreviation is UPID (Universal Portfolio Identity), and it is empty that the space where them is known as universe investment combination
Between.And conventional investment theory, due to the limitation of computing capability at that time, the investment combination amount of space analyzed is well below this
Quantity.Therefore it needs to regard with broader from above-mentioned universe investment combination space using state-of-the-art big data technology
Analysis A share market is gone at angle, establishes the machine learning system for being directed to super flood tide Portfolio Selection, and make its efficiency and income meeting
Considerably beyond previous traditional quantization investment analysis.
In novel quantization investment field, quantitative analysis teacher be typically from personal share, by statistical method or
Machine learning is predicted, selects stock.These methods overcome the shortcomings that conventional investment field, but this side to a certain extent
The combination that method is produced not is global optimum's combination.
Invention content
In view of this, the purpose of the present invention is to provide a kind of Portfolio Selection selection method, device and equipment, to throw
Money is combined as research object, is selected from all possible investment combination and generates optimal Portfolio.
In a first aspect, an embodiment of the present invention provides a kind of Portfolio Selection selection method, including machine learning model,
Method includes：Obtain the history feature value of history investment combination；History feature value is inputted into initial machine learning model, is obtained close
Phase earning rate predicted value；The difference of recent earning rate predicted value and recent earning rate actual value is calculated, determination obtains minimal difference
When corresponding model parameter；Model parameter is inputted into initial machine learning model, obtains optimization machine learning model；By current spy
Value indicative input optimization machine learning model, obtains next period earning rate predicted value, Portfolio Selection selection is carried out for user.
With reference to first aspect, an embodiment of the present invention provides the first possible embodiments of first aspect, wherein obtains
The step of taking the history feature value of history investment combination, including：History investment combination is screened according to pre-defined rule, is obtained
Target investment combination；The characteristic value data in target investment combination is obtained, characteristic value data includes that technical indicator, basic side refer to
Mark, statistical indicator and spectrum signature；Characteristic value data is handled and is calculated, history feature value is obtained.
With reference to first aspect and its first possible embodiment, an embodiment of the present invention provides the second of first aspect
The possible embodiment of kind, wherein history investment combination is screened according to pre-defined rule, obtains the step of target investment combination
Suddenly, including：Selection meets the history investment combination of pre-defined rule, as investment combination to be selected；It is chosen from investment combination to be selected
The lower target investment combination of personal share correlation.
With reference to first aspect, an embodiment of the present invention provides the third possible embodiments of first aspect, wherein meter
The difference of recent earning rate predicted value and recent earning rate actual value is calculated, determination obtains corresponding model parameter when minimal difference
Step, including：The recent earning rate that corresponding investment combination is predicted by history feature value, to obtain recent earning rate predicted value；
It obtains history feature and is worth corresponding investment combination in recent earning rate actual value；By recent earning rate predicted value and recent income
The difference of rate actual value obtains corresponding model parameter when target function value minimum as object function.
With reference to first aspect, an embodiment of the present invention provides the 4th kind of possible embodiments of first aspect, wherein will
Current characteristic value input optimization machine learning model, obtains next period earning rate predicted value, Portfolio Selection choosing is carried out for user
The step of selecting, including：Next period earning rate predicted value is conveyed to display terminal；The screening inputted by display terminal according to user
Output with conditions Portfolio Selection selection result；Shown selection result is supplied to user.
Second aspect, the embodiment of the present invention also provide a kind of Portfolio Selection selection device, using machine learning model,
Including：History feature value acquisition module, the history feature value for obtaining history investment combination；Computing module is used for history
Characteristic value inputs initial machine learning model, obtains recent earning rate predicted value；Training module, it is pre- for calculating recent earning rate
The difference of measured value and recent earning rate actual value, determination obtain corresponding model parameter when minimal difference；Optimization module, being used for will
Model parameter inputs initial machine learning model, obtains optimization machine learning model；Interactive module, for current characteristic value is defeated
Enter to optimize machine learning model, obtain next period earning rate predicted value, Portfolio Selection selection is carried out for user.
In conjunction with second aspect, an embodiment of the present invention provides the first possible embodiment of second aspect, history is special
Value indicative acquisition module is additionally operable to:History investment combination is screened according to pre-defined rule, obtains target investment combination；Obtain mesh
The characteristic value data in investment combination is marked, characteristic value data includes that technical indicator, basic side index, statistical indicator and frequency spectrum are special
Sign；Characteristic value data is handled and is calculated, history feature value is obtained.
In conjunction with second aspect and its first possible embodiment, an embodiment of the present invention provides the second of second aspect
The possible embodiment of kind, device history feature value acquisition module are additionally operable to：Selection meets the history investment combination of pre-defined rule,
As investment combination to be selected；The lower target investment combination of personal share correlation is chosen from investment combination to be selected.
In conjunction with second aspect, an embodiment of the present invention provides the third possible embodiments of second aspect, wherein hands over
Mutual module is additionally operable to：Next period earning rate predicted value is conveyed to display terminal；The screening inputted by display terminal according to user
Output with conditions Portfolio Selection selection result；Shown selection result is supplied to user.
The third aspect, the embodiment of the present invention also provide a kind of electronic equipment, including memory, processor, are deposited in memory
The computer program that can be run on a processor is contained, processor realizes above-mentioned first aspect when executing computer program and its can
Any one embodiment of energy.
The embodiment of the present invention brings following advantageous effect：Portfolio Selection selecting party provided in an embodiment of the present invention
Method, device and equipment get the history feature value of history investment combination, pass through engineering using investment combination as research object
Model is practised, the relationship between research history characteristic value and historical return predicts recent earning rate, then by minimizing last-period forecast
Difference between earning rate and recent effective yield, optimizes machine learning model, finally by current characteristic value and optimization
Machine learning model produce the predicted value of next period earning rate, selected as desired for user.It is single with investment combination
Position carries out the selection of Portfolio Selection, can select optimal Portfolio from all possible investment combination.
Other feature and advantage of the disclosure will illustrate in the following description or Partial Feature and advantage can be from
Specification deduces or unambiguously determines, or by implement the disclosure above-mentioned technology it can be learnt that.
To enable the above objects, features, and advantages of the disclosure to be clearer and more comprehensible, preferred embodiment cited below particularly, and coordinate
Appended attached drawing, is described in detail below.
Description of the drawings
It, below will be to specific in order to illustrate more clearly of the specific embodiment of the invention or technical solution in the prior art
Embodiment or attached drawing needed to be used in the description of the prior art are briefly described, it should be apparent that, in being described below
Attached drawing is some embodiments of the present invention, for those of ordinary skill in the art, before not making the creative labor
It puts, other drawings may also be obtained based on these drawings.
Fig. 1 is Portfolio Selection selection method flow chart provided in an embodiment of the present invention；
Fig. 2 is the flow of model parameter preparation method in Portfolio Selection selection method provided in an embodiment of the present invention
Figure；
Fig. 3 is the flow of selection result preparation method in Portfolio Selection selection method provided in an embodiment of the present invention
Figure；
Fig. 4 is the structure diagram of Portfolio Selection selection device provided in an embodiment of the present invention；
Fig. 5 is that Portfolio Selection provided in an embodiment of the present invention selects electronic devices structure block diagram.
Icon：
41- history feature value acquisition modules；42- computing modules；43- training modules；44- optimization modules；45- interacts mould
Block；51- memories；52- processors.
Specific implementation mode
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with attached drawing to the present invention
Technical solution be clearly and completely described, it is clear that described embodiments are some of the embodiments of the present invention, rather than
Whole embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art are not making creative work premise
Lower obtained every other embodiment, shall fall within the protection scope of the present invention.
It is well known that the investment target in financial market is numerous.With economic and finance development, investment target more they tends to more
Sample.For A share market, a total of more than 2800 stock of A share market at present, then being demonstrate,proved caused by this 2800 stocks
Just there are about as many as 10 33 powers for certificate investment combination quantity.The investment combination of this 10 33 power is referred to as universe investment combination,
English abbreviation is UPID (Universal Portfolio Identity), and it is empty that the space where them is known as universe investment combination
Between.And conventional investment theory, due to the limitation of computing capability at that time, the investment combination amount of space analyzed is well below this
Quantity.Therefore it needs to regard with broader from above-mentioned universe investment combination space using state-of-the-art big data technology
Analysis A share market is gone at angle, establishes the machine learning system for being directed to super flood tide Portfolio Selection, and make its efficiency and income meeting
Considerably beyond previous traditional quantization investment analysis.
In novel quantization investment field, quantitative analysis teacher be typically from personal share, by statistical method or
Machine learning is predicted, selects stock.These methods overcome the shortcomings that conventional investment field, but this side to a certain extent
The combination that method is produced not is global optimum's combination.Based on this, a kind of Portfolio Selection provided in an embodiment of the present invention
Selection method, device and equipment are selected from all possible investment combination and are generated optimal using investment combination as research object
Investment combination.
For ease of understanding the present embodiment, a kind of Portfolio Selection disclosed in the embodiment of the present invention is selected first
Selection method describes in detail.
Embodiment 1
The embodiment of the present invention 1 provides a kind of Portfolio Selection selection method, Portfolio Selection shown in Figure 1
The flow chart and selection result preparation method shown in Fig. 3 of selection method flow chart and model parameter preparation method shown in Fig. 2
Flow chart, this approach includes the following steps：
Step S102 obtains the history feature value of history investment combination.
At the T moment, the historical juncture is divided into three points, t by we0、t1、t2, wherein t0<t1<t2<T.We obtain t0It arrives
t1The characteristic value of the investment combination of period, referred to as the history feature value of history investment combination.
In step s 102, the step of obtaining the history feature value of history investment combination, including：History investment combination is pressed
It is screened according to pre-defined rule, obtains target investment combination；Obtain the characteristic value data in target investment combination, characteristic value data
Including technical indicator, basic side index, statistical indicator and spectrum signature；Characteristic value data is handled and is calculated, is gone through
History characteristic value.
Characteristic value data is the 1*m vectors by being obtained in target investment combination, and m is characterized number, such as m=8+5+8+7
=28, the summation of as 8 technical indicators and 5 basic side indexs and 8 statistical indicators and 7 spectrum signatures.Technology is joined
Number includes simple Moving Average, weighted moving average line, index Moving Average, smooth unusual fluctuation Moving Average, the rate of change
Index, inertia index, relative strength index, basic side index include scale beta, market beta, value/growth beta, the country
Total output value factor is open, price index factor is open, and statistical indicator includes standard deviation, the degree of bias, kurtosis, Jarque-Bera inspections
Test, Jarque-Bera is examined, KPSS is examined, Dickey-Fuller GLS are examined, Hurst indexes, spectrum signature includes energy
Mutual information, kurtosis, harmonic wave index, frequency domain center of gravity, frequency variance, time domain center of gravity, comentropy.Characteristic value data is with the lattice of floating number
Formula stores.
Characteristic value data is handled and is calculated, obtaining history feature value includes：Stop in removal target investment combination
The corresponding data of board stock, to be cleaned to target investment combination data；To removing power and/or ex dividend in target investment combination
Stock certificate data multiple power processing before carrying out, to carry out noise to target investment combination data；Characteristic value data is based on distribution
Storage, and calculating is normalized based on large-scale data processing frame, to obtain characteristic value.
In step s 102, history investment combination is screened according to pre-defined rule, obtains the step of target investment combination
Suddenly, including：Selection meets the history investment combination of pre-defined rule, as investment combination to be selected；It is chosen from investment combination to be selected
The lower target investment combination of personal share correlation.
The relatively high investment combination of historical yield is chosen from investment combination database, then screens wherein personal share correlation ratio
Lower investment combination is as target investment combination.Pre-set rule can be above specific earning rate.For example, setting in advance
The rule set can be：Investment combination of the earning rate 5% or more then chooses earning rate 5% from investment combination database
Above investment combination.
The related coefficient of all personal share yield volatilities between any two in investment combination is calculated, i.e.,：
Wherein, X is the yield volatility of stock 1, and Y is the yield volatility of stock 2.Calculate the phase of all investment combinations
Guan Xinghou chooses the lower investment combination of correlation as target investment combination.
History feature value is inputted initial machine learning model, obtains recent earning rate predicted value by step S104.
Obtain t0To t1The history feature value of period inputs machine learning model, is given birth to according to the setting of machine mould itself
It is t in the recent period by decision function prediction at decision function1To t2The earning rate of period：Our given input x, x, that is, history are special
Value indicative, model can export a predicted value fθ(x), wherein θ is model parameter, decision function fθConcrete form by machine mould
The setting decision of itself, such as in perceptron machine learning model,Different machines learning model
Decision function is different.
Step S106, calculates the difference of recent earning rate predicted value and recent earning rate actual value, and determination obtains lowest difference
Corresponding model parameter when value.
Obtain i.e. t in the recent period1To t2The actual value y of period earning rate, by the predicted value f of recent earning rateθ(x) and it is practical
The difference of value y is as object function：J (θ)=(y-fθ(x))2, object function is minimized by gradient descent method, obtains corresponding mould
Shape parameter.Model parameter include algorithm the convergence speed, the weight of characteristic value and model complexity punishment parameter.
In step s 106, the difference of recent earning rate predicted value and recent earning rate actual value is calculated, determination obtains most
When small difference the step of corresponding model parameter, including：
Step S202 predicts the recent earning rate of corresponding investment combination by history feature value, to obtain recent earning rate
Predicted value.
Step S204 obtains history feature and is worth corresponding investment combination in recent earning rate actual value.
Step S206 is obtained using the difference of recent earning rate predicted value and recent earning rate actual value as object function
Corresponding model parameter when target function value minimum.
Model parameter is inputted initial machine learning model by step S108, obtains optimization machine learning model.
Current characteristic value is inputted optimization machine learning model, next period earning rate predicted value is obtained, for user by step S110
Carry out Portfolio Selection selection.
Investment combination is obtained in current i.e. t2To the characteristic value of T time section income, this feature value is inputted into optimization engineering
Practise model, the prediction receipts of the machine learning model of optimization can provide the next period based on the model parameter after optimization i.e. T to T+c periods
Beneficial rate.By this predicted value, i.e. next period indicated yield is supplied to user, user that can be selected by brushing as the final output of model
The expected yield of each investment combination chooses the investment combination oneself admired.
In step s 110, current characteristic value is inputted into optimization machine learning model, obtains next period earning rate predicted value, supplied
User carries out the step of Portfolio Selection selection, including：
Next period earning rate predicted value is conveyed to display terminal by step S302.
Display terminal can be app software interfaces or webpage, for being interacted for user and system.
Step S304 exports Portfolio Selection selection result according to user by the screening conditions that display terminal inputs.
User can display terminal input screening conditions, screening conditions can be expected rate, risk partiality or
Custom parameter.Custom parameter can be the easy number of days of most short delivery, most with respect to deep bid winning rate, history average absolute income, history
Withdraw greatly, history Sharpe Ratio, include the listed company of combination are averaged annual net profit amplification etc., user can be by showing end
End inputs one or more screening conditions, and system can filter investment combination database according to the demand of client, be accorded with
Conjunction condition simplifies investment combination.Such as user can set in 250 day of trade, using every 7 day of trade as sliding window, often
The earning rate of 7 day of trade is above the investment combination of Shanghai and Shenzhen 300.Programmer is given by inputting the parameter in python scripts
Go out through all investment combinations after this condition filter.
User's expected revenus is inputted by user oneself it is expected year earning rate section, such as 5%-10%, 10%-15% etc.,
System can filter out investment combination of the desired year earning rate between user input area in existing investment combination database, and
It is supplied to user；System can weigh the risk partiality of user by questionnaire form, while calculating it to all investment combinations and going through
History stability bandwidth is simultaneously divided into low-risk, risk and high risk three parts, and corresponding investment is provided according to the risk partiality of user
Combination.It can be it is recommended that under the identification record for the investment combination crossed, when recommending investment combination next time in daily recommend
Ensure that the mark of the investment combination of user is supplied to not appear in the same day it has been recommended that list in, to ensure it is mutual not
It repeats.Simultaneously in view of price shock be many users while while buying and selling same stock generate to cause share price violent
The transaction cost generated is fluctuated, since we are supplied to the investment combination of user not repeat mutually, is less likely to occur very
The case where multiple users buy and sell same stock simultaneously occurs, to reduce the reality for improving combination for share price price shock
Income.
Shown selection result is supplied to user by step S306.
User receives final selection result by client.
The embodiment of the present invention gets the history feature value of history investment combination, passes through using investment combination as research object
Machine learning model, the relationship between research history characteristic value and historical return predict recent earning rate, then close by minimizing
Difference between phase indicated yield and recent effective yield, optimizes machine learning model, finally by current characteristic value
The predicted value that next period earning rate is produced with the machine learning model of optimization selects as desired for user.With investment group
Unit is combined into carry out the selection of Portfolio Selection, Optimal Investment group can be selected from all possible investment combination
It closes.
Embodiment 2
The embodiment of the present invention 2 provides a kind of Portfolio Selection selection device, Portfolio Selection shown in Figure 4
The structure diagram of selection device, the device include：History feature value acquisition module 41, the history for obtaining history investment combination
Characteristic value；Computing module 42 obtains recent earning rate predicted value for history feature value to be inputted initial machine learning model；
Training module 43, the difference for calculating recent earning rate predicted value and recent earning rate actual value, determination obtain minimal difference
When corresponding model parameter；Optimization module 44 obtains optimization engineering for model parameter to be inputted initial machine learning model
Practise model；Interactive module 45 optimizes machine learning model for inputting current characteristic value, obtains next period earning rate predicted value,
Portfolio Selection selection is carried out for user.
History feature value acquisition module 41 is additionally operable to:History investment combination is screened according to pre-defined rule, obtains mesh
Mark investment combination；Obtain target investment combination in characteristic value data, characteristic value data include technical indicator, basic side index,
Statistical indicator and spectrum signature；Characteristic value data is handled and is calculated, history feature value is obtained.
History feature value acquisition module 41 is additionally operable to：Selection meets the history investment combination of pre-defined rule, as throwing to be selected
Money combination；The lower target investment combination of personal share correlation is chosen from investment combination to be selected.
Interactive module 45 is additionally operable to：Next period earning rate predicted value is conveyed to display terminal；According to user by showing eventually
The screening conditions of end input export Portfolio Selection selection result；Selection result is supplied to user.
The technique effect and preceding method embodiment phase of the device that the embodiment of the present invention is provided, realization principle and generation
Together, to briefly describe, device embodiment part does not refer to place, can refer to corresponding contents in preceding method embodiment.
Device provided in an embodiment of the present invention, the method technical characteristic having the same provided with above-described embodiment, so
Also identical technical problem can be solved, identical technique effect is reached.
Embodiment 3
The embodiment of the present invention 3 provides a kind of electronic equipment, electronic devices structure block diagram shown in Figure 5, including deposits
Reservoir 51, processor 52 are stored with the computer program that can be run on the processor 52 in memory 51, and processor 52 executes meter
The method that arbitrary steps describe in embodiment 1 is realized when calculation machine program.
It is apparent to those skilled in the art that for convenience and simplicity of description, the equipment of foregoing description
Specific work process, can refer to corresponding processes in the foregoing method embodiment, details are not described herein.
Unless specifically stated otherwise, the opposite step of the component and step that otherwise illustrate in these embodiments, digital table
It is not limit the scope of the invention up to formula and numerical value.
In all examples being illustrated and described herein, any occurrence should be construed as merely illustrative, without
It is as limitation, therefore, other examples of exemplary embodiment can have different values.
Finally it should be noted that：Embodiment described above, only specific implementation mode of the invention, to illustrate the present invention
Technical solution, rather than its limitations, scope of protection of the present invention is not limited thereto, although with reference to the foregoing embodiments to this hair
It is bright to be described in detail, it will be understood by those of ordinary skill in the art that：Any one skilled in the art
In the technical scope disclosed by the present invention, it can still modify to the technical solution recorded in previous embodiment or can be light
It is readily conceivable that variation or equivalent replacement of some of the technical features；And these modifications, variation or replacement, do not make
The essence of corresponding technical solution is detached from the spirit and scope of technical solution of the embodiment of the present invention, should all cover the protection in the present invention
Within the scope of.Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (10)
1. a kind of Portfolio Selection selection method, which is characterized in that including machine learning model, the method includes：
Obtain the history feature value of history investment combination；
The history feature value is inputted into initial machine learning model, obtains recent earning rate predicted value；
It calculates the difference of the recent earning rate predicted value and recent earning rate actual value, determines when obtaining the minimum difference pair
The model parameter answered；
The model parameter is inputted into the initial machine learning model, obtains optimization machine learning model；
Current characteristic value is inputted into the optimization machine learning model, obtains next period earning rate predicted value, security are carried out for user
Portfolio Selection Based.
2. according to the method described in claim 1, it is characterized in that, the step of the history feature value for obtaining history investment combination
Suddenly, including：
The history investment combination is screened according to pre-defined rule, obtains target investment combination；
Obtain the characteristic value data in the target investment combination, the characteristic value data include technical indicator, basic side index,
Statistical indicator and spectrum signature；
The characteristic value data is handled and calculated, the history feature value is obtained.
3. according to the method described in claim 2, it is characterized in that, described sieve history investment combination according to pre-defined rule
The step of selecting, obtaining target investment combination, including：
Selection meets the history investment combination of pre-defined rule, as investment combination to be selected；
The lower target investment combination of personal share correlation is chosen from the investment combination to be selected.
4. according to the method described in claim 1, it is characterized in that, described calculate the recent earning rate predicted value and receive in the recent period
The difference of beneficial rate actual value, determining the step of obtaining corresponding model parameter when the minimum difference, including：
The recent earning rate that corresponding investment combination is predicted by history feature value, to obtain recent earning rate predicted value；
It obtains the history feature and is worth corresponding investment combination in the recent earning rate actual value；
Using the difference of the recent earning rate predicted value and recent earning rate actual value as object function, the target letter is obtained
Corresponding model parameter when numerical value minimum.
5. according to the method described in claim 1, it is characterized in that, described input the optimization machine learning by current characteristic value
Model obtains next period earning rate predicted value, for user carry out Portfolio Selection selection the step of, including：
The next period earning rate predicted value is conveyed to display terminal；
According to user Portfolio Selection selection result is exported by the screening conditions that the display terminal inputs；
Shown selection result is supplied to user.
6. a kind of Portfolio Selection selection device, which is characterized in that using machine learning model, described device includes：
History feature value acquisition module, the history feature value for obtaining history investment combination；
Computing module obtains recent earning rate predicted value for the history feature value to be inputted initial machine learning model；
Training module, the difference for calculating the recent earning rate predicted value and recent earning rate actual value, determination obtain most
Corresponding model parameter when the small difference；
Optimization module obtains optimization machine learning model for the model parameter to be inputted the initial machine learning model；
Interactive module obtains next period earning rate predicted value, supplies for current characteristic value to be inputted the optimization machine learning model
User carries out Portfolio Selection selection.
7. device according to claim 6, which is characterized in that the history feature value acquisition module is additionally operable to:
History investment combination is screened according to pre-defined rule, obtains target investment combination；Obtain the target investment combination
In characteristic value data, the characteristic value data include technical indicator, basic side index, statistical indicator and spectrum signature；
The characteristic value data is handled and calculated, history feature value is obtained.
8. device according to claim 7, which is characterized in that the history feature value acquisition module is additionally operable to：
Selection meets the history investment combination of pre-defined rule, as investment combination to be selected；
The lower target investment combination of personal share correlation is chosen from the investment combination to be selected.
9. device according to claim 6, which is characterized in that the interactive module is additionally operable to：
The next period earning rate predicted value is conveyed to display terminal；
According to user Portfolio Selection selection result is exported by the screening conditions that the display terminal inputs；
Shown selection result is supplied to user.
10. a kind of electronic equipment, including memory, processor, it is stored with and can runs on the processor in the memory
Computer program, which is characterized in that the processor realizes that the claims 1-5 is arbitrary when executing the computer program
Method described in one.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810242799.3A CN108492189A (en) | 2018-03-22 | 2018-03-22 | Portfolio Selection selection method, device and equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810242799.3A CN108492189A (en) | 2018-03-22 | 2018-03-22 | Portfolio Selection selection method, device and equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108492189A true CN108492189A (en) | 2018-09-04 |
Family
ID=63319518
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810242799.3A Pending CN108492189A (en) | 2018-03-22 | 2018-03-22 | Portfolio Selection selection method, device and equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108492189A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110852888A (en) * | 2019-10-18 | 2020-02-28 | 浙江工业大学 | Particle filter-based security investment portfolio optimization method |
| CN111242311A (en) * | 2020-01-06 | 2020-06-05 | 高盈量化云科技（深圳）有限公司 | Multi-line set effective output method |
| CN111311420A (en) * | 2020-02-21 | 2020-06-19 | 深圳市思迪信息技术股份有限公司 | Business data pushing method and device |
| CN111861711A (en) * | 2020-07-22 | 2020-10-30 | 未鲲(上海)科技服务有限公司 | Resource allocation method and related product |
| CN112862620A (en) * | 2021-03-31 | 2021-05-28 | 山东大学 | Investment product combination recommendation method and system based on investor preference |
-
2018
- 2018-03-22 CN CN201810242799.3A patent/CN108492189A/en active Pending
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110852888A (en) * | 2019-10-18 | 2020-02-28 | 浙江工业大学 | Particle filter-based security investment portfolio optimization method |
| CN111242311A (en) * | 2020-01-06 | 2020-06-05 | 高盈量化云科技（深圳）有限公司 | Multi-line set effective output method |
| CN111311420A (en) * | 2020-02-21 | 2020-06-19 | 深圳市思迪信息技术股份有限公司 | Business data pushing method and device |
| CN111861711A (en) * | 2020-07-22 | 2020-10-30 | 未鲲(上海)科技服务有限公司 | Resource allocation method and related product |
| CN111861711B (en) * | 2020-07-22 | 2024-06-07 | 重庆百盐投资(集团)有限公司 | Resource allocation method and related product |
| CN112862620A (en) * | 2021-03-31 | 2021-05-28 | 山东大学 | Investment product combination recommendation method and system based on investor preference |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108492189A (en) | Portfolio Selection selection method, device and equipment | |
| Qin et al. | Blockchain: a carbon-neutral facilitator or an environmental destroyer? | |
| CN109784779B (en) | Financial risk prediction method, device and storage medium | |
| Lia et al. | Cost Minimization is Essential for the Sustainable Development of an Industry: A Mathematical Economic Model Approach | |
| CN108320053A (en) | A kind of region electricity demand forecasting method, apparatus and system | |
| CN102496126B (en) | Custody asset transaction data monitoring equipment | |
| Maji et al. | Stock prediction and mutual fund portfolio management using curve fitting techniques | |
| CN108229750A (en) | A kind of stock yield Forecasting Methodology | |
| CN1971610A (en) | System and method for evaluating bank lending risks | |
| Song et al. | The impacts of nonferrous metal price shocks on the macroeconomy in China from the perspective of resource security | |
| CN108492190A (en) | Intelligent security Portfolio Selection Based method, apparatus and equipment | |
| Zhou et al. | Supplier’s goal setting considering sustainability: An uncertain dynamic Data Envelopment Analysis based benchmarking model | |
| Yi et al. | Financial risk prediction in supply chain finance based on buyer transaction behavior | |
| Zhu et al. | Measuring the performance of international trade using a DEA-based approach with trade imbalances consideration | |
| CN104321800A (en) | Price target builder | |
| CN109919647A (en) | The dependency prediction system and method for financial products | |
| CN108985595A (en) | The move transaction service evaluation method and device mutually commented based on counterparty | |
| CN107578130A (en) | A kind of method of the copper forecasting of futures prix of the TSK fuzzy systems of structure and parameter combination learning | |
| CN107480703A (en) | Transaction fault detection method and device | |
| Peachavanish | Dual time frame relative strength stock selection using fuzzy logic | |
| Shil et al. | Meeting peak electricity demand through combinatorial reverse auctioning of renewable energy | |
| TWI248007B (en) | Method for evaluating market trade based on trend prediction | |
| CN109598623A (en) | A kind of financial product future profits data predication method, apparatus and system | |
| CN108846761A (en) | A kind of option futures exchange analysis method based on artificial neural network | |
| Zandi et al. | Research on stock portfolio based on time series prediction and multi-objective optimization |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180904 |