CN109493139A - Similar market judgement, preferred method and storage medium - Google Patents
Similar market judgement, preferred method and storage medium Download PDFInfo
- Publication number
- CN109493139A CN109493139A CN201811323482.9A CN201811323482A CN109493139A CN 109493139 A CN109493139 A CN 109493139A CN 201811323482 A CN201811323482 A CN 201811323482A CN 109493139 A CN109493139 A CN 109493139A
- Authority
- CN
- China
- Prior art keywords
- amount
- information
- decrease
- increase
- opposite
- 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
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0202—Market predictions or forecasting for commercial activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/04—Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange
Abstract
A kind of similar market judgement, preferred method and storage medium, wherein similar market judgment method includes the following steps, computing unit obtains historical transactional information；Obtain the first Transaction Information to be predicted, first Transaction Information to be predicted is identical as the amount of increase and amount of decrease information node number of historical transactional information, the opposite amount of increase and amount of decrease information of each node in historical transactional information or the first Transaction Information to be predicted is calculated, opposite amount of increase and amount of decrease information is amount of increase and amount of decrease information of each node with respect to first node in Transaction Information sequence；Historical transactional information is calculated at a distance from the opposite amount of increase and amount of decrease of corresponding node in the first Transaction Information to be predicted, the distance of the opposite amount of increase and amount of decrease for all corresponding nodes of summing.The present invention can be convenient for reference over market, provide the solution of current quotations calculating similar to historical quotes, for selection by the user.
Description
Technical field
The present invention relates to financial product design field more particularly to a kind of forecast for market tendency of the financial product based on similarity
Method.
Background technique
The Object of Transaction of trade market is vast as the open sea at present, by taking stock as an example, domestic market can transaction's stock have it is thousands of
Personal share, friend stock invester can not accurately filter out the target admired when operating transaction, so as to cause looking at flowers in a fog, hope in water
Month, it selects stocks and misses the main points.There is a kind of idea of simplicity always for a long time, if it is possible to pass through similar market, it will be able to pre-
Survey the similar tendency of After-market, that is to say, that, it is generally recognized that same market can repeat to show, but see the heart only according to human eye and be only difficult
It is similar market that, which is analyzed, need a kind of science, objective, quantifiable index carry out auxiliary judgment.So exist
In the exchange quotation APP that friend stock invester uses, it is necessary to integrate a kind of functional method, the historical trading in database can be obtained
Information can receive user to the selection instruction of market, carry out corresponding scientific algorithm, science judgment, science and compare, finally to
User shows most preferred individual company share quotations information.
Summary of the invention
In order to rationally quantify more similar market, it is desirable to provide one kind can be convenient for reference over market, provide and work as
The solution of preceding market calculating similar to historical quotes for selection by the user, while promoting the user experience of market class application
With brand competitiveness.
To achieve the above object, inventor has developed a kind of similar market judgment method, includes the following steps, computing unit
Obtain historical transactional information；
Obtain the first Transaction Information to be predicted, the amount of increase and amount of decrease letter of first Transaction Information to be predicted and historical transactional information
Breath node number is identical, calculates the opposite amount of increase and amount of decrease information of each node in historical transactional information or the first Transaction Information to be predicted,
Opposite amount of increase and amount of decrease information is amount of increase and amount of decrease information of each node with respect to first node in Transaction Information sequence；
Historical transactional information is calculated at a distance from the opposite amount of increase and amount of decrease of corresponding node in the first Transaction Information to be predicted, summation
The distance of the opposite amount of increase and amount of decrease of all corresponding nodes.
Further, Transaction Information further includes closing price information or amount of increase and amount of decrease information, further includes step,
It is calculated by the closing price information or amount of increase and amount of decrease information of each node and obtains opposite amount of increase and amount of decrease information.
It specifically, further include step, the distance of the opposite amount of increase and amount of decrease of all corresponding nodes of weighted sum, time series is more early
Corresponding node opposite amount of increase and amount of decrease distance weight it is lower.
Preferably, the Transaction Information further includes exchange hand information, further includes step, calculates historical transactional information or first
The opposite exchange hand amount of increase and amount of decrease information of each node in transaction to be predicted；
Calculate the opposite exchange hand amount of increase and amount of decrease of corresponding node in historical transactional information and the first Transaction Information to be predicted away from
From the distance of the opposite exchange hand amount of increase and amount of decrease for all corresponding nodes of summing；
Further include step, comprehensive similarity distance is calculated, by the opposite amount of increase and amount of decrease of all corresponding nodes apart from summed result
The summed result weighted sum at a distance from opposite exchange hand amount of increase and amount of decrease with all corresponding nodes obtains comprehensive similarity distance.
A kind of similar market preferred method, including step, user terminal, which receives user, believes the selection of historical transactional information
Breath obtains the number of nodes of the historical transactional information of selection, the transaction in default screening range is chosen, from when day before yesterday forward trace phase
Same node carries out similar market judgement as the first Transaction Information to be predicted, the Transaction Information include closing price information or at
Friendship amount information；
The opposite amount of increase and amount of decrease information of each node in calculating historical transactional information, the first Transaction Information to be predicted；Calculate history
Transaction Information is at a distance from the opposite amount of increase and amount of decrease of corresponding node in the first Transaction Information to be predicted, the phase for all corresponding nodes of summing
To the distance of amount of increase and amount of decrease；Or calculate historical transactional information, in the first Transaction Information to be predicted each node opposite exchange hand ups and downs
Width information, calculate the opposite exchange hand amount of increase and amount of decrease of corresponding node in historical transactional information and the first Transaction Information to be predicted away from
From the distance of the opposite exchange hand amount of increase and amount of decrease for all corresponding nodes of summing.
Further, further include step, comprehensive similarity distance is calculated, by the distance of the opposite amount of increase and amount of decrease of all corresponding nodes
Summed result summed result weighted sum at a distance from opposite exchange hand amount of increase and amount of decrease with all corresponding nodes, obtain integrating it is similar away from
From.
Specifically, the similarity of each Transaction Information to be predicted and historical transactional information is calculated according to comprehensive similarity distance, and
It is pushed to user and shows the similarity, the similarity of each Transaction Information to be predicted indicates are as follows: 1 and personal share total distance calculated result
Account for the difference of the ratio of the total distance maximum value in default screening range.
A kind of similar market judge storage medium, are stored with computer program, the computer program when executed into
Row includes the following steps that computing unit obtains historical transactional information；
Obtain the first Transaction Information to be predicted, the amount of increase and amount of decrease letter of first Transaction Information to be predicted and historical transactional information
Breath node number is identical, calculates the opposite amount of increase and amount of decrease information of each node in historical transactional information or the first Transaction Information to be predicted,
Opposite amount of increase and amount of decrease information is amount of increase and amount of decrease information of each node with respect to first node in Transaction Information sequence；
Historical transactional information is calculated at a distance from the opposite amount of increase and amount of decrease of corresponding node in the first Transaction Information to be predicted, summation
The distance of the opposite amount of increase and amount of decrease of all corresponding nodes.
Further,
Transaction Information further includes closing price information or amount of increase and amount of decrease information, and the computer program also carries out such as when executed
Lower step,
It is calculated by the closing price information or amount of increase and amount of decrease information of each node and obtains opposite amount of increase and amount of decrease information,
The distance of the opposite amount of increase and amount of decrease of all corresponding nodes of weighted sum, the opposite of the more early corresponding node of time series are risen
The weight of the distance of drop range is lower；
It also carries out including step, the opposite exchange hand for calculating each node in historical transactional information or the first transaction to be predicted rises
Drop range information；
Calculate the opposite exchange hand amount of increase and amount of decrease of corresponding node in historical transactional information and the first Transaction Information to be predicted away from
From the distance of the opposite exchange hand amount of increase and amount of decrease for all corresponding nodes of summing；
Comprehensive similarity distance is calculated, by the opposite amount of increase and amount of decrease of all corresponding nodes apart from summed result and all corresponding sections
Point opposite exchange hand amount of increase and amount of decrease apart from summed result weighted sum, obtain integrate similarity distance.
A kind of preferred medium of similar market, is stored with computer program, and the computer program is wrapped when executed
Step is included,
User terminal receives user to the selection information of historical transactional information, obtains the node of the historical transactional information of selection
Number chooses the transaction in default screening range, from when day before yesterday forward trace same node point, as the first Transaction Information to be predicted into
The similar market judgement of row, the Transaction Information includes closing price information or exchange hand information；
The opposite amount of increase and amount of decrease information of each node in calculating historical transactional information, the first Transaction Information to be predicted；Calculate history
Transaction Information is at a distance from the opposite amount of increase and amount of decrease of corresponding node in the first Transaction Information to be predicted, the phase for all corresponding nodes of summing
To the distance of amount of increase and amount of decrease；Or calculate historical transactional information, in the first Transaction Information to be predicted each node opposite exchange hand ups and downs
Width information, calculate the opposite exchange hand amount of increase and amount of decrease of corresponding node in historical transactional information and the first Transaction Information to be predicted away from
From the distance of the opposite exchange hand amount of increase and amount of decrease for all corresponding nodes of summing.
It is different from the prior art, above scheme can be by way of standardization price or amount of increase and amount of decrease sequence, and quantization is different
Transaction Information in corresponding similarity degree, can be used for show or as the sort by etc. for selecting Object of Transaction is attained, always away from
From bigger, then the similarity of two K lines is lower, and total distance is smaller, and the similarity of two K lines is higher, such displaying result phase
It is more intuitive compared with directly range estimation K line chart, the data for needing to be compared calculating can also be chosen depending on the user's operation
Source, therefore user experience is also stronger, increases the product viscosity of user, enhances the competitiveness that Transaction Information shows product.
Detailed description of the invention
Fig. 1 is similar market judgment method flow chart described in the specific embodiment of the invention.
Specific embodiment
Technology contents, construction feature, the objects and the effects for detailed description technical solution, below in conjunction with specific reality
It applies example and attached drawing is cooperated to be explained in detail.
In the embodiment shown in fig. 1, a kind of similar market judgment method is illustrated, this method can be implemented in user's
Handheld terminal can be equipped with corresponding exchange quotation in the handheld terminal and show APP, can integrate in APP program for holding
The computing module of the row operation, such as computing unit.And in the case of other, the above method can also be implemented in server end or
In the integrated program module of database side.Specifically, this method may begin at following steps, and S104 obtains historical trading letter
Breath, the Transaction Information includes amount of increase and amount of decrease information；
S106 obtains the first Transaction Information to be predicted, the ups and downs of first Transaction Information to be predicted and historical transactional information
Width information number is identical, calculates the opposite amount of increase and amount of decrease information of each node in historical transactional information or the first Transaction Information to be predicted；
S108 calculates historical transactional information at a distance from the opposite amount of increase and amount of decrease of corresponding node in the first Transaction Information to be predicted,
The distance of the opposite amount of increase and amount of decrease for all corresponding nodes of summing.The calculation of the distance includes but is not limited to (1) Euclidean distance:
Current most widely used metric range, it can be readily appreciated that realizing simple.The disadvantage is that, the price variance of different stocks is larger, meeting
It, can be to avoid this problem with standardized method normalizing even if form is similar between the stock for causing price difference big.(2) dynamic
Time Warp distance (DTW): it is widely used in field of speech recognition, it is based on dynamic programming principle.Disadvantage is
Temporal influence is not embodied, general recent form tendency more has predictability than form tendency at a specified future date.(3) longest
Common subsequence distance (LCS): based on longest common subsequence distance with longest subsequence identical in two time serieses
The ratio of length and length of time series measures the similitude between two time serieses.Handling discontinuous sequence timeliness
Fruit is better than two kinds of front distance, and disadvantage needs to be standardized to reduce error as Euclidean distance.In the present solution,
Our distance is preferably Euclidean distance, and such calculation amount is small, as a result also relatively precisely.
In our embodiment, historical transactional information can be read by database, and transaction here is that can buy and sell throwing
Provide target, such as stock, bond, futures, real estate market, collectibles.The node is a certain statistics moment, node and a upper node
Between price change can be defined as the amount of increase and amount of decrease information that this programme will be discussed, can be day, ten between node and node
Day, the moon, hour, 30 minutes etc. statistical intervals.By taking stock exchange as an example, we choose day line and are used as explanation, then amount of increase and amount of decrease information
It is exactly the relative price ups and downs on the same day Yu the previous day of trade, certainly in some cases, Transaction Information includes closing price information, I
Can pass through closing price information and calculate and obtain amount of increase and amount of decrease information or opposite amount of increase and amount of decrease information by step.It should be noted that
Opposite amount of increase and amount of decrease information is amount of increase and amount of decrease information of each node with respect to first node in Transaction Information sequence.Opposite amount of increase and amount of decrease can be with
It is obtained by the calculating of each node closing price information, that is to say, that after either knowing closing price or amount of increase and amount of decrease information
Opposite amount of increase and amount of decrease is all easy to acquire, and opposite amount of increase and amount of decrease can be good at being standardized as index, ignore personal share
Price bring influences.
For example, continuous three days prices are 100 yuan, 110 yuan, 121 yuan in certain historical transactional information；Amount of increase and amount of decrease be 0,
10%, 10%；Opposite amount of increase and amount of decrease is 0,10%, 21%；And another continuous three days prices of Transaction Information are 10 yuan, 11 yuan, 12.1
The stock of member has same data appearance after the standardized concepts processing of opposite amount of increase and amount of decrease.In some other standardized application
In example, calculating can also be standardized by following formula:
Wherein x is the node data in sequence, and x is the average value of sequence, and S is standard deviation, and x ' is the node after standardization
Data value can play the role of excluding the influence of price absolute value, the standardization formula is under as above-mentioned opposite amount of increase and amount of decrease
It is equally applicable in the standardization of numerical value such as the exchange hand of text.
Thus in a still further embodiment, if the closing price of two sections of isometric K line number evidences of stock A and stock B is not
Pa1,Pa2,…PanAnd Pb1,Pb2,…Pbn, exchange hand Va1,Va2,…VanAnd Vb1,Vb2,…Vbn
Then amount of increase and amount of decrease Zi=(Pi-Pi-1)/Pi-1, opposite amount of increase and amount of decrease Ai=(Pi-P1)/P1, i is node in place transaction data
In precedence.The distance of the opposite amount of increase and amount of decrease of corresponding node is determined by following formula: distAi=(Aai-Abi)2。
The finally total distance of two K lines of summation,, DabFor price similarity；Through the above steps,
It can quantify corresponding similarity degree in different Transaction Informations, can be used for showing or as attaining the sequence for selecting Object of Transaction
According to etc., total distance is bigger, then the similarity of two K lines is lower, and total distance is smaller, and the similarity of two K lines is higher, in this way
Displaying result compared to directly range estimation K line chart it is more intuitive, user experience is also stronger, increase the product viscosity of user, increase
Strong Transaction Information shows the competitiveness of product.
In some embodiments, the Transaction Information further includes exchange hand information, further includes step, and S110 calculates history and hands over
The opposite exchange hand amount of increase and amount of decrease information of each node in easy information or the first transaction to be predicted.Calculate historical transactional information with first to
Predict that the distance of the opposite exchange hand amount of increase and amount of decrease of corresponding node in Transaction Information, the opposite exchange hand for all corresponding nodes of summing rise
The distance of drop range.Specifically, if two sections of isometric K line numbers of above-mentioned stock A and stock B according to exchange hand be Va1,Va2,…VanAnd Vb1,
Vb2,…Vbn；Corresponding exchange hand amount of increase and amount of decrease, opposite exchange hand amount of increase and amount of decrease here relatively, which can also be calculated, is, subsequent node
Exchange hand was relative to first day quantity ups and downs changing tendency, and i-th of sequence node is with respect to exchange hand amount of increase and amount of decrease Mi=(Vi-V1)/
V1, with respect to the interference that exchange hand amount of increase and amount of decrease can exclude exchange hand numerical value itself, can more objectively respond between several nodes
Exchange hand result of variations.Then the distance calculating method between stock A and stock B with respect to exchange hand amount of increase and amount of decrease is distMi=
(Mai-Mbi)2.Finally sum to the distance of exchange hand amount of increase and amount of decreaseEabFor exchange hand similarity.By upper
Step is stated, the similarity degree of corresponding conclusion of the business figureofmerit in different Transaction Informations can be quantified, can be used for showing or is made
To attain the sort by etc. for selecting Object of Transaction, total distance is bigger, then the similarity of two K lines is lower, and total distance is smaller, and two
The similarity of K line is higher, and such displaying result is more intuitive compared to directly range estimation exchange hand, and user experience is also stronger, increases
The product viscosity for having added user enhances the competitiveness that Transaction Information shows product.
Designer has found in practical applications, and the K line of same price tendency still has larger difference on subsequent performance,
Concurrently show exchange hand has a large effect in the comparison of data, and amount valence, which deviates from, the forms such as to be risen also with amount valence together there is research to anticipate
Justice.Therefore in certain embodiments, in order to promoted similarity judgement confidence level and method stability.Further include step, calculates
Comprehensive similarity distance, by striking a bargain apart from summed result and the opposite of all corresponding nodes for the opposite amount of increase and amount of decrease of all corresponding nodes
Measure amount of increase and amount of decrease apart from summed result weighted sum, obtain comprehensive similarity distance Qab。
Qab=wp×Dab+wv×Eab
By comprehensive similarity distance, to avoid valence amount problem not in the same direction.It is calculated by share price merely, Ke Nengzao
It is similar at form, but actual market environment and personal share conclusion of the business situation difference are huge, and reference significance is not strong.By being counted simultaneously to valence amount
It calculates, weighted accumulation obtains synthesis similarity distance to the end, increases the validity of matching result, has preferably reached different transaction letters
The effect that similarity determines between breath.
In some other further embodiment, influence of the recent market market to price series validity should be greater than remote
Influence of the phase market conditions to price series validity is based on this, then in the opposite amount of increase and amount of decrease of all corresponding nodes of weighted sum
Distance, or opposite exchange hand amount of increase and amount of decrease distance makes, the power of the distance of the opposite amount of increase and amount of decrease of the more early corresponding node of time series
Again lower, the influence to total distance is then just smaller, and weight can pass through the logarithm letter of ordinal number of the node in entire data sequence
The modes such as number, exponential function, power function or trigonometric function obtain.It tries by taking logarithmic function as an example, then total distance DabAnd EabIt can be with table
It is shown as:
By above-mentioned optimization, we have better played the technical effect of similarity degree between two groups of Transaction Informations of evaluation.
And in some other preferred embodiment, it is based on above-mentioned method for evaluating similarity system, we can also use
Historical transactional information matching feature is selected at family, as shown in Figure 1, the present invention program further includes step, S100 receives user to history
The selection information of Transaction Information obtains the number of nodes of the historical transactional information of selection, chooses any friendship in default screening range
Easily, from Transaction Information that the number of nodes is included when day before yesterday forward trace same node point number, is extracted as the first transaction to be predicted
Information carries out similar market judgement.Specifically, it is assumed that user clicks similar market matching feature in application interface, has chosen 1 year
Before rise sharply certain stock K line of preceding 30 day of trade, after our application obtains the selection result, record historical trading letter
Breath, and total node number is judged for 30, default screening range can be stock markets of Shanghai stock, Shenzhen stock market stock, middle platelet stock, GEM
The screening range that stock or user's self-selected stock etc. pre-set, then our method will traverse all personal shares in range, with most
New Transaction day is the Transaction Information that starting point chooses forward 30 nodes, is successively gone through as the first Transaction Information to be predicted with what is chosen
History Transaction Information carries out similitude comparison, show that total distance calculated result information, total distance here can be opposite amount of increase and amount of decrease
Total distance, be also possible to the total distance of opposite exchange hand amount of increase and amount of decrease, be defaulted as comprehensive total distance, therefore can also include step
Suddenly, the selection that user compares preference to Transaction Information is received, including compares price, compare exchange hand or global alignment.Finally
Comparison result in addition to directly pushing the personal share that distance is minimum in default screening range be that other than preferred result is shown, can also carry out
Step: S112 calculates the similarity of preferred result, and pushes to user and show the similarity, and the similarity of each personal share indicates
Are as follows: 1 accounts for the difference of the ratio of the total distance maximum value in default screening range, i.e. similarity with personal share total distance calculated result
Sim
Wherein DIST is the total distance calculated result of certain branch personal share and historical transactional information, can be DabOr EabOr Qab, m
For the set of all personal shares in default screening range, DIST [m] is the maximum total distance calculated result in the set.
It should be noted that being not intended to limit although the various embodiments described above have been described herein
Scope of patent protection of the invention.Therefore, it based on innovative idea of the invention, change that embodiment described herein is carried out and is repaired
Change, or using equivalent structure or equivalent flow shift made by description of the invention and accompanying drawing content, it directly or indirectly will be with
Upper technical solution is used in other related technical areas, is included within scope of patent protection of the invention.
Claims (10)
1. a kind of similar market judgment method, which is characterized in that include the following steps, computing unit obtains historical transactional information；
Obtain the first Transaction Information to be predicted, the amount of increase and amount of decrease information section of first Transaction Information to be predicted and historical transactional information
Point number is identical, calculates the opposite amount of increase and amount of decrease information of each node in historical transactional information or the first Transaction Information to be predicted, relatively
Amount of increase and amount of decrease information is amount of increase and amount of decrease information of each node with respect to first node in Transaction Information sequence；
Historical transactional information is calculated at a distance from the opposite amount of increase and amount of decrease of corresponding node in the first Transaction Information to be predicted, summation is all
The distance of the opposite amount of increase and amount of decrease of corresponding node.
2. similar market judgment method according to claim 1, which is characterized in that Transaction Information further includes closing price information
Or amount of increase and amount of decrease information, it further include step,
It is calculated by the closing price information or amount of increase and amount of decrease information of each node and obtains opposite amount of increase and amount of decrease information.
3. similar market judgment method according to claim 1, which is characterized in that further include step, weighted sum is all
The distance of the opposite amount of increase and amount of decrease of corresponding node, the weight of the distance of the opposite amount of increase and amount of decrease of the more early corresponding node of time series is more
It is low.
4. similar market judgment method according to claim 1, which is characterized in that the Transaction Information further includes exchange hand
Information further includes step, calculates the opposite exchange hand amount of increase and amount of decrease letter of each node in historical transactional information or the first transaction to be predicted
Breath；
Historical transactional information is calculated at a distance from the opposite exchange hand amount of increase and amount of decrease of corresponding node in the first Transaction Information to be predicted, is asked
The distance of opposite exchange hand amount of increase and amount of decrease with all corresponding nodes；
Further include step, comprehensive similarity distance is calculated, by the opposite amount of increase and amount of decrease of all corresponding nodes apart from summed result and institute
Have the opposite exchange hand amount of increase and amount of decrease of corresponding node apart from summed result weighted sum, obtain comprehensive similarity distance.
5. a kind of similar market preferred method, which is characterized in that including step, user terminal receives user to historical transactional information
Selection information, obtain the number of nodes of the historical transactional information of selection, choose the transaction in default screening range, from when the day before yesterday to
Preceding backtracking same node point carries out similar market judgement as the first Transaction Information to be predicted, and the Transaction Information includes closing price
Information or exchange hand information；
The opposite amount of increase and amount of decrease information of each node in calculating historical transactional information, the first Transaction Information to be predicted；Calculate historical trading
Information at a distance from the opposite amount of increase and amount of decrease of corresponding node in the first Transaction Information to be predicted, rise by the opposite of all corresponding nodes of summing
The distance of drop range；Or the opposite exchange hand amount of increase and amount of decrease of each node is believed in calculating historical transactional information, the first Transaction Information to be predicted
Breath calculates historical transactional information at a distance from the opposite exchange hand amount of increase and amount of decrease of corresponding node in the first Transaction Information to be predicted, asks
The distance of opposite exchange hand amount of increase and amount of decrease with all corresponding nodes.
6. similar market preferred method according to claim 5, which is characterized in that further include step, it is similar to calculate synthesis
Distance, by the opposite exchange hand amount of increase and amount of decrease apart from summed result and all corresponding nodes of the opposite amount of increase and amount of decrease of all corresponding nodes
Apart from summed result weighted sum, obtain comprehensive similarity distance.
7. similar market preferred method according to claim 6, which is characterized in that according to comprehensive similarity distance calculate respectively to
It predicts the similarity of Transaction Information and historical transactional information, and is pushed to user and show the similarity, each transaction letter to be predicted
The similarity of breath indicates are as follows: 1 accounts for the ratio of the total distance maximum value in default screening range with personal share total distance calculated result
Difference.
8. a kind of similar market judge storage medium, which is characterized in that be stored with computer program, the computer program is in quilt
Included the following steps when execution, computing unit obtains historical transactional information；
Obtain the first Transaction Information to be predicted, the amount of increase and amount of decrease information section of first Transaction Information to be predicted and historical transactional information
Point number is identical, calculates the opposite amount of increase and amount of decrease information of each node in historical transactional information or the first Transaction Information to be predicted, relatively
Amount of increase and amount of decrease information is amount of increase and amount of decrease information of each node with respect to first node in Transaction Information sequence；
Historical transactional information is calculated at a distance from the opposite amount of increase and amount of decrease of corresponding node in the first Transaction Information to be predicted, summation is all
The distance of the opposite amount of increase and amount of decrease of corresponding node.
9. a kind of similar market according to claim 8 judge storage medium, which is characterized in that
Transaction Information further includes closing price information or amount of increase and amount of decrease information, and the computer program is also walked as follows when executed
Suddenly,
It is calculated by the closing price information or amount of increase and amount of decrease information of each node and obtains opposite amount of increase and amount of decrease information,
The distance of the opposite amount of increase and amount of decrease of all corresponding nodes of weighted sum, the opposite amount of increase and amount of decrease of the more early corresponding node of time series
Distance weight it is lower；
It also carries out including step, calculates the opposite exchange hand amount of increase and amount of decrease of each node in historical transactional information or the first transaction to be predicted
Information；
Historical transactional information is calculated at a distance from the opposite exchange hand amount of increase and amount of decrease of corresponding node in the first Transaction Information to be predicted, is asked
The distance of opposite exchange hand amount of increase and amount of decrease with all corresponding nodes；
Comprehensive similarity distance is calculated, by the opposite amount of increase and amount of decrease of all corresponding nodes apart from summed result and all corresponding nodes
With respect to exchange hand amount of increase and amount of decrease apart from summed result weighted sum, comprehensive similarity distance is obtained.
10. a kind of preferred medium of similar market, which is characterized in that be stored with computer program, the computer program is being held
It carries out including step when row,
User terminal receives user to the selection information of historical transactional information, obtains the number of nodes of the historical transactional information of selection,
The transaction in default screening range is chosen, from when day before yesterday forward trace same node point, is carried out as the first Transaction Information to be predicted
Similar market judgement, the Transaction Information includes closing price information or exchange hand information；
The opposite amount of increase and amount of decrease information of each node in calculating historical transactional information, the first Transaction Information to be predicted；Calculate historical trading
Information at a distance from the opposite amount of increase and amount of decrease of corresponding node in the first Transaction Information to be predicted, rise by the opposite of all corresponding nodes of summing
The distance of drop range；Or the opposite exchange hand amount of increase and amount of decrease of each node is believed in calculating historical transactional information, the first Transaction Information to be predicted
Breath calculates historical transactional information at a distance from the opposite exchange hand amount of increase and amount of decrease of corresponding node in the first Transaction Information to be predicted, asks
The distance of opposite exchange hand amount of increase and amount of decrease with all corresponding nodes.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811323482.9A CN109493139A (en) | 2018-11-08 | 2018-11-08 | Similar market judgement, preferred method and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811323482.9A CN109493139A (en) | 2018-11-08 | 2018-11-08 | Similar market judgement, preferred method and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109493139A true CN109493139A (en) | 2019-03-19 |
Family
ID=65695394
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811323482.9A Pending CN109493139A (en) | 2018-11-08 | 2018-11-08 | Similar market judgement, preferred method and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109493139A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111192144A (en) * | 2020-01-03 | 2020-05-22 | 湖南工商大学 | Financial data prediction method, device, equipment and storage medium |
-
2018
- 2018-11-08 CN CN201811323482.9A patent/CN109493139A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111192144A (en) * | 2020-01-03 | 2020-05-22 | 湖南工商大学 | Financial data prediction method, device, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Li | Sparse and stable portfolio selection with parameter uncertainty | |
| US8352347B2 (en) | Investment classification and tracking system using diamond ratings | |
| WO2015048181A1 (en) | Systems and methods for presenting relevant data to users of a financial computer network | |
| Das* | A learning market-maker in the Glosten–Milgrom model | |
| US20140279693A1 (en) | Goal-Based Portfolio Management System | |
| KR102509348B1 (en) | Method and device for providing stock recommendation service | |
| Bayrakdaroğlu et al. | Strategic financial performance evaluation of the Turkish companies traded on ISE | |
| US20180322580A1 (en) | Intelligent Financial Management System | |
| Meligkotsidou et al. | Quantile forecast combinations in realised volatility prediction | |
| Martens et al. | Carry investing on the yield curve | |
| DK202370110A1 (en) | Environmental, social, and governance (esg) performance trends | |
| Debata et al. | Interdependence between monetary policy and stock liquidity: A panel VAR approach | |
| Kim et al. | Time series regression-based pairs trading in the Korean equities market | |
| Henriques et al. | A multiobjective interval portfolio framework for supporting investor’s preferences under different risk assumptions | |
| WO2012075488A1 (en) | Private company valuation | |
| CN107665383A (en) | A kind of construction method and computing device of resume processing delay model | |
| Evgeniou et al. | Uncovering sparsity and heterogeneity in firm-level return predictability using machine learning | |
| CN109493139A (en) | Similar market judgement, preferred method and storage medium | |
| KR20160028701A (en) | System for selecting stock using buying and selling index | |
| Stasinakis et al. | Krill-Herd Support Vector Regression and heterogeneous autoregressive leverage: evidence from forecasting and trading commodities | |
| WO2001003046A1 (en) | A method and system to synthesize portfolios of goods, services or financial instruments | |
| Liu et al. | Non-homogeneous volatility correlations in the bivariate multifractal model | |
| Kim et al. | A closer look at return predictability of the US stock market: evidence from new panel variance ratio tests | |
| Swartz et al. | Relative Value Hedge Funds: A behavioral modeling of hedge fund risk and return factors | |
| Wang et al. | Skilled mutual fund selection: false discovery control under dependence |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190319 |