CN112734577B - Multi-period decision AI (advanced technology attachment) strand-frying robot platform - Google Patents
Multi-period decision AI (advanced technology attachment) strand-frying robot platform Download PDFInfo
- Publication number
- CN112734577B CN112734577B CN202110223333.0A CN202110223333A CN112734577B CN 112734577 B CN112734577 B CN 112734577B CN 202110223333 A CN202110223333 A CN 202110223333A CN 112734577 B CN112734577 B CN 112734577B
- Authority
- CN
- China
- Prior art keywords
- transaction
- module
- stock
- trading
- line
- 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.)
- Active
Links
- 238000012216 screening Methods 0.000 claims abstract description 56
- 238000013500 data storage Methods 0.000 claims abstract description 19
- 230000000007 visual effect Effects 0.000 claims abstract description 7
- 238000001914 filtration Methods 0.000 claims abstract description 3
- 230000000630 rising effect Effects 0.000 claims description 64
- 238000004458 analytical method Methods 0.000 claims description 52
- 230000008859 change Effects 0.000 claims description 14
- 230000004913 activation Effects 0.000 claims description 13
- 230000003213 activating effect Effects 0.000 claims description 9
- 238000000034 method Methods 0.000 claims description 8
- 238000012790 confirmation Methods 0.000 claims description 6
- 230000007547 defect Effects 0.000 claims description 5
- 241000282376 Panthera tigris Species 0.000 claims description 3
- 230000002159 abnormal effect Effects 0.000 claims description 3
- 230000001364 causal effect Effects 0.000 claims description 2
- 239000003086 colorant Substances 0.000 claims description 2
- 230000010485 coping Effects 0.000 abstract description 3
- 239000000284 extract Substances 0.000 abstract description 2
- 230000007774 longterm Effects 0.000 description 9
- 230000000877 morphologic effect Effects 0.000 description 8
- 230000008901 benefit Effects 0.000 description 7
- 238000013473 artificial intelligence Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 6
- 230000006870 function Effects 0.000 description 6
- 238000013139 quantization Methods 0.000 description 6
- 230000008961 swelling Effects 0.000 description 6
- 238000013461 design Methods 0.000 description 4
- 230000001960 triggered effect Effects 0.000 description 4
- 230000001976 improved effect Effects 0.000 description 3
- 230000001149 cognitive effect Effects 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 230000008451 emotion Effects 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000005096 rolling process Methods 0.000 description 2
- 230000035939 shock Effects 0.000 description 2
- 208000001613 Gambling Diseases 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000001934 delay Effects 0.000 description 1
- 230000009189 diving Effects 0.000 description 1
- 230000005284 excitation Effects 0.000 description 1
- 230000003116 impacting effect Effects 0.000 description 1
- 230000006698 induction Effects 0.000 description 1
- 230000001939 inductive effect Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 210000001503 joint Anatomy 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- TYJJADVDDVDEDZ-UHFFFAOYSA-M potassium hydrogencarbonate Chemical compound [K+].OC([O-])=O TYJJADVDDVDEDZ-UHFFFAOYSA-M 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
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
- G06Q40/04—Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Finance (AREA)
- General Physics & Mathematics (AREA)
- Accounting & Taxation (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- General Engineering & Computer Science (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Technology Law (AREA)
- General Business, Economics & Management (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The invention discloses a multi-period decision AI (automatic teller machine) stock-frying robot platform, which adopts a 64-bit k line combination chart classification coding and step-by-step billing type data storage technology, uses big data statistics to decompose and classify nuances of stock price trend characteristics, writes code names, intuitively reflects trend forms of various types of stock price trend in a tracking index mode, extracts an optimal quantitative transaction model from the trend forms for screening stocks, guides preferred stocks into a stock pool of a transaction target after screening and filtering, and finally activates the stocks through transaction instructions to realize full-automatic transaction of the AI stock-frying robot. The visual quantitative decision tool provided by the invention helps a user to improve instant coping capability and complex decision capability, is convenient to use, can greatly improve investment efficiency, rapidly and accurately activates stocks of a trade target to finish trade, and obtains ultrahigh yield which cannot be obtained by manual stock frying.
Description
Technical Field
The invention relates to a multi-period decision AI (automatic identification) stranding robot platform, and belongs to the technical field of artificial intelligent trading platforms.
Background
The common defects of the prior market analysis systems such as stocks, futures and derivatives are that the k-line diagram, the average line index, the fluctuation index and the trend index are separated from the basic surface, the internal structure of stock price trend cannot be clearly represented, the buying and selling signals are lagged, a high-winning decision scheme is difficult to be rapidly given in the transient market, investors are often caused to fall into blind rising and falling gambling traps, and the investment risk is increased.
In the application aspect of AI (automatic identification) stranding robots, a multi-factor quantization model is generally adopted, k lines with high rising probability and high yield are manufactured into a quantization model, and stock screening and trading are performed after a plurality of technical face index factors and basic face financial factors are added. The so-called multi-factor full-automatic transaction quantization model does not realize real-time butt joint of a basic surface and a technical surface, often falls into decision dilemma of a great deal of return annual income ratio and a great deal of real income ratio, and the quantization transaction model is difficult to cope with market change with unusual change.
Disclosure of Invention
The invention aims to overcome the defects in the prior art, provides a multi-period decision AI (automatic identification) strand-frying robot platform, solves the problem of splitting of a basic surface and a technical surface, realizes the organic combination of basic surface information flow driving and technical surface fund flow driving, returns to a sunny rain gauge function of a stock market, correctly reveals the change rule of stock price, and increases the accuracy of comprehensive analysis and prediction; the method solves the problem of partial probability of k-line diagram and index analysis, overcomes the defects of linear thinking and perceptual decision, attaches importance to the internal structure of stock price trend, deeply reveals the essential rule of stock price fluctuation, forms a more rational strategy scheme by using a multi-dimensional and multi-level comprehensive analysis method, and improves the operation accuracy; the problem of the signal hysteresis of buying and selling is solved, the trend characteristic of quotation is prompted in real time, the transaction command is automatically triggered when the requirement of a preset algorithm is met, the loss risk of signal time lag is solved, market change is timely dealt with, and the accuracy and timeliness of operation are improved.
In order to achieve the above purpose, the invention adopts the technical means that: a multi-period decision AI (automatic input) stock-frying robot platform adopts a 64-bit k-line combined graph classification coding and step-by-step billing type data storage technology, a big data statistics method is used for decomposing and classifying nuances of stock price trend characteristics, code names are written, trend forms of various trend types and stock prices are intuitively reflected in a mode of tracking indexes, an optimal quantitative transaction model is extracted from the trend forms and used for screening stocks, the optimal stocks are imported into a stock pool of a transaction target after screening and filtering, and finally the full-automatic transaction of the AI stock-frying robot is realized through transaction instruction activation.
Further, the full-automatic transaction platform of the multi-period decision AI stranding robot comprises a platform server, a transaction platform sub-machine, wherein a plurality of transaction platform sub-machines are connected with the platform server, a transaction data acquisition module (1), a 64-bit k line combination classification coding module (2), a time-sharing k line combination graph code analysis module (3), a time-sharing transaction model determination module (4), an information acquisition module (5), a 64-bit k line combination graph code classification step-by-step accounting type data storage module (6), a multi-period k line combination graph code index module (7), a multi-period k line combination graph code analysis module (8), a multi-period transaction model determination module (9) and a transaction model screening stock module (10) are arranged in the platform server;
The output end of the transaction data acquisition module (1) is connected with the input end of the 64-bit k-line combination classification coding module (2), transaction data of the abnormal stocks are acquired, abstract data are provided for billing storage, and correct data support is provided for selection of a transaction model;
The output end of the 64-bit k-line combination classification coding module (2) is connected with the input end of the time-sharing k-line combination graph code analysis module (3), and the 64-bit k-line combination classification coding module (2) codes 6 k-line combination graphs into 64 k-line combination codes according to the rising and falling characteristics according to coding rules, so as to define 64 trend types; according to the rising and falling amplitude characteristics of each k line, a four-color k line graph is compiled, 64-bit k line combination codes are expanded into 4096 k line combination codes, 4096 trend types are defined, and therefore fine classification of stock price trend types is achieved; the output end of the time-sharing k line combined graph code analysis module (3) is connected with the input end of the time-sharing transaction model determination module (4), and the output end of the time-sharing transaction model determination module (4) is connected with the input end of the transaction model screening stock module (10);
The time-sharing k line combination pattern code analysis module (3) selects the trend type of the time-sharing k line combination of the current day with high rising and falling probability of the next day from the output data of the 64-bit k line combination classification coding module (2), and the time-sharing transaction model determination module (4) determines the trend type of the time-sharing k line combination of the current day with high rising and falling probability of the next day as a real-time transaction model;
The output end of the information acquisition module (5) is connected with the input end of the 64-bit k-line combined image code classified step-by-step billing type data storage module (6), the output end of the 64-bit k-line combined image code classified step-by-step billing type data storage module (6) is connected with the input end of the multi-period k-line combined image code index module (7), the information acquisition module (5) acquires basic surface information influencing the change of the stock price in real time, defines the properties of emptiness and interest, and is respectively placed into large disc information, industry information, regional information and individual stock information according to the causal relations of the influence of the stock price, the content of the marked by marking symbols with different colors is distinguished by the properties of the interest and the interest, the content of the marking column is used for reflecting the inherent relation of the change of the stock price, the objective of analysis prediction is increased, and correct information support is provided for the selection of a transaction model;
The 64-bit k-line combined graph code classification step-by-step billing data storage module (6) sets 64 classification daily accounts and 4096 detail classification daily accounts according to a 64-bit k-line combined coding principle, sets transaction daily accounts of each stock correspondingly, and records basic face information, transaction amount, expansion and fall values and expansion and fall amplitudes into corresponding daily accounts after each transaction day is finished; after the end of each week of transaction day, the transaction amount, the rising and falling value and the rising and falling value are transferred into corresponding week classification accounts; after the transaction day at the end of each month, the transaction amount, the rise and fall value and the rise and fall amplitude are transferred into corresponding month classification accounts;
The output end of the multi-period k-line combined picture code index module (7) is connected with the input end of the multi-period k-line combined picture code analysis module (8), the multi-period k-line combined picture code index module (7) compiles a stock price type tracking index by using k-line combined picture codes, the rising frequency and the income rate of the previous month k, the previous week k and the next month, the next week and the next day adjacent to each other of the previous day k-line combined picture codes are arranged from high to low, so that the rising and falling frequency and the income rate trend of subdivision trend types are more visual, and quantifiable rising and falling probability and income rate data are provided for analysis decisions of users; the multi-period k-line combined graph code analysis module (8) selects a transaction model of a front month, a front week and a front day k-line combined graph code of a stock price trend type which is most easy to rise or fall in a later period through index tracking data and basic surface information data;
The output end of the multi-period k-line combined graphic code analysis module (8) is connected with the input end of the multi-period transaction model determination module (9), and the output end of the multi-period transaction model determination module (9) is connected with the input end of the transaction model screening stock module (10); the multi-period trading model determining module (9) determines a quantitative trading model for buying or selling according to the trend type defined by the k line combination graph codes of the previous month, the previous week and the previous day, wherein the rising probability of the next period is high and the yield is high or the falling probability of the next period is high and the loss rate is high, and the trading model screening stock module (10) filters the trading models input by the multi-period trading model determining module and the time-sharing trading module step by step and screens the trading models as trading targets in real time according to the tracking index and the back measurement analysis of the trend type.
Further, the four-color k line graph is formed, namely: the small-rise K-line, the purple-color K-line, the green-color K-line, and the blue-color K-line are indicated by red, green, and blue, respectively.
Further, the transaction data acquisition module acquires transaction data of the stock, including: the stock exchange period, the time length, the expansion and drop amplitude, the hand change rate, the large bill transaction amount, the expansion and drop stop board, the institution transaction data published by the dragon and tiger list and various index data.
Further, the summary columns of the journal are two columns, the two columns record basic surface information and transaction data information respectively, and classified and step-by-step accounting data are stored as big data for artificial intelligence analysis.
Further, the transaction platform sub-machine (12) comprises a transaction model screening stock selection module, a basic surface factor screening stock selection module, a transaction target stock pool module, a transaction instruction setting module, a transaction instruction ticket reporting module, a transaction instruction activating module and a transaction completion confirmation module;
The transaction model screening stock selection module receives data of a platform server, compiles and selects k line combined graph codes of month, week, day and time sharing period, and the transaction model is set by a user according to the transaction idea and style of the user;
The basic surface factor screening stock selection module receives transaction model information of the transaction model screening stock selection module, a user sets the selected basic surface factors according to own transaction concepts and styles, and information factors, market rates, dynamic market rates, market net rates, total stakes, total market values and circulation market values are selected;
The stock pool module of the trade target receives the selected data of the basic surface factor screening stock selection module, and performs real-time screening to form the stock pool of the trade target;
The trade command setting module is connected with the trade target stock pool module, selects stocks from the trade target stock pool, sets trade funds, the number and the amount of the stocks, activates the trading command and provides a preset command for full-automatic trade;
The transaction instruction newspaper module receives the data information of the transaction instruction setting module, and automatically matches the number, the quantity and the amount of stocks according to a preset transaction rule and a buying and selling direction to form newspaper information;
The transaction instruction activation module receives the newspaper information of the transaction instruction newspaper module, sends the newspaper information into the platform server, the platform server automatically executes the transaction instruction according to the buying and selling directions according to the preset transaction rule, when a certain target stock in the newspaper information meets the preset algorithm condition, the platform server sends out the transaction instruction to the transaction instruction activation module, and when a certain target stock in the newspaper information does not meet the preset algorithm condition, the platform server sends out the transaction instruction to the transaction instruction activation module;
The confirmation completion transaction module receives an activation transaction instruction of the transaction instruction activation module, purchases or sells a certain target stock in newspaper information meeting the buying or selling conditions, after the transaction is completed, a transaction result is transmitted to the platform server, after the transaction is completed, the target stock pool module of the transaction stores the purchased stock into a selling column of a target stock pool of the transaction, and the transaction waits for full-automatic selling transaction; for the stocks that have completed the sell trade, the trade target stock pool module deletes them from the sell column of the trade target stock pool;
The platform server (11) controls the trading of the trading platform sub-machine (12), a user sets a trading model, technical indexes, information factors and basic factors through the trading platform sub-machine (12) to form a trading strategy scheme, and the trading platform sub-machine (12) automatically executes, screens stocks and carries out full-automatic trading.
Further, in the stock pool of the trade target stock pool module, the user observes the real-time dynamic of the trade target stock and allows the stock to be added and removed, and finally the stock of the trade target stock pool is formed and becomes the full-automatic trade target stock.
The invention has the beneficial effects that:
1. A visual comprehensive quantitative decision tool is provided for a user, and the user is helped to promote instant coping capability and complex decision capability. The high integration of the basic plane information and the technical plane trend is realized by adopting the classified step-by-step billing type data storage, and the correlation between the basic plane information and the technical plane share price trend is objectively reflected; furthermore, the tracking index of the k line combined graph codes of the subdivision trend types is adopted, so that the later period rising and falling frequency and the yield curve of each trend type are more visual; and by utilizing an artificial intelligence big data processing technology, the stock price trend type integrated by basic surface information and technical surfaces is subjected to mathematical statistics analysis, so that a user is helped to promote instant coping capacity and complex decision making capacity, and ultra-high income is obtained.
2. Convenient use and greatly improved investment efficiency. The method is convenient for a user to select, arrange and combine long, medium, short and ultra-short-term strategies, and various and multi-period quantized investment strategies can be formed by only adding and deleting the combination chart codes of the month k, the week k, the day k and the time-sharing k lines; then, by adding and deleting the basic plane multifactor, a comprehensive transaction strategy scheme meeting the investment preference and the wind control requirement of the user can be prepared; the platform automatically selects stocks according to the set trading strategy and generates a target stock pool of the corresponding investment strategy at a high speed, so that the workload that manual analysis of stock selection can be completed within a few minutes only by a few months can be completed, investment errors caused by personal emotion fluctuation and cognitive deviation are reduced, and stock selection time of a user is greatly saved.
3. Can accurately activate the stock of the trade target to finish the trade at high speed, and obtain the ultrahigh yield which can not be obtained by manually stir-frying the stock. The stock frying robot utilizes the advantages of an artificial intelligent algorithm, greatly expands the stock selection visual angle, automatically searches and screens all stocks, helps a user to capture rising opportunities of individual stocks at high speed among instant matters, reduces time lags and delays generated by personal observation, judgment and manual report, accurately and efficiently completes transactions, greatly improves the transaction efficiency, and effectively configures funds into the stocks which are most prone to rising, thereby acquiring investment advantages which are incomparable with manual analysis and manual report, and realizing maximization of income.
4. The manual operation can be completely separated from manual operation, and the excess rolling profit and excess rebate benefits are realized. In a default transaction model of the platform or in a transaction model selected by a user, automatic searching can be realized to conduct full-automatic transaction, so that long-term stable rolling profit and excess complex profit are obtained.
Drawings
The invention is further illustrated in the following figures and examples.
FIG. 1 is a block diagram of a multi-cycle decision AI stock-frying robot system of the present invention;
FIG. 2 is a block diagram of a full-automatic trading platform system according to the present invention;
FIG. 3 is a schematic diagram of a k-line combined graph classification encoding system according to the present invention;
In the figure: 1. the system comprises a transaction data acquisition module, a 2-64-bit k-line combination classification coding module, a 3-time-sharing k-line combination graph code analysis module, a 4-time-sharing transaction model determination module, a 5-information acquisition module, a 6-64-bit k-line combination graph code classification step-by-step billing type data storage module, a 7-multicycle k-line combination graph code index module, an 8-multicycle k-line combination graph code analysis module, a 9-multicycle transaction model determination module, a 10-transaction model screening stock module, an 11-platform server, a 12-transaction platform sub-machine.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The full-automatic trading platform of the multi-period decision AI (automatic identification) stranding robot shown in figures 1,2 and 3 comprises a platform server, a trading platform sub-machine, wherein a plurality of trading platform sub-machines are connected with one platform server, and a trading data acquisition module 1, a 64-bit k-line combination classification coding module 2, a time-sharing k-line combination pattern code analysis module 3, a time-sharing trading model determination module 4, an information acquisition module 5, a 64-bit k-line combination pattern code classification step-by-step accounting type data storage module 6, a multi-period k-line combination pattern code index module 7, a multi-period k-line combination pattern code analysis module 8, a multi-period trading model determination module 9 and a trading model screening stock module 10 are arranged in the trading platform sub-machine;
The output end of the transaction data acquisition module 1 is connected with the input end of the 64-bit k-line combination classification coding module 2, transaction data of the abnormal stocks are acquired, abstract data are provided for billing storage, and correct data support is provided for selection of a transaction model;
The output end of the 64-bit k-line combination classification coding module 2 is connected with the input end of the time-sharing k-line combination graph code analysis module 3, and the 64-bit k-line combination classification coding module 2 codes 6 k-line combination graphs into 64 k-line combination codes according to the rising and falling characteristics and defines 64 trend types; according to the rising and falling amplitude characteristics of each k line, a four-color k line graph is compiled, 64-bit k line combination codes are expanded into 4096 k line combination codes, 4096 trend types are defined, fine classification of stock price trend types is achieved, and support is provided for an visualized, identifiable and interpretable big data technology analysis system;
The 64-bit k-line combined classification coding module is actually a method for completely classifying stock price trends, which is used for finely classifying and coding all the stock price trends of each stock in a trade period, and defining the trend types, and takes the intra-day time sharing 64-bit k-line combined classification coding as an example, as shown in fig. 3: one trade day is 240 minutes, and six k lines in 40-minute class are decomposed to be determined as y; each k line is decomposed into three 13-minute (namely 13 x 3 approximately 40) secondary level k lines, and the x is determined; y, calculating the total fluctuation range of the three x, coding by using positive value of rise and negative value of fall, (represented by red k line when y is larger than 0, the single code of the encodable letter is A, represented by green k line when y is smaller than or equal to 0, the single code of the encodable letter is B), thus generating 64 groups of codes by using AB two letters and defining 64 trend types for 6 y combinations; y calculates the total rising and falling amplitude of the three x, encodes the rising and falling amplitude with positive value and negative value, and distinguishes the rising and falling amplitude, (when 1% > y is more than 0, the encodable letter single code is represented by red k line, when y is more than 1%, the encodable letter single code is represented by purple k line, when 0 > y is more than-1%, the encodable letter single code is represented by green k line, the encodable letter single code is represented by b, when-1% > y, the encodable letter single code is represented by blue k line, so that 4096 groups of codes can be generated for 6 y combinations with abcd four letters, and 4096 trend types are defined; in total 262626144 k line combinations on the order of 13 minutes, in one transaction day, (i.e. 18 x, represented by the two k-lines of yin and yang, the combination is 18 times 2 = 2 x 2 = 2626144), the 64-bit classifier effectively classifies and codes 26262820 combinations of 13 minutes in a day into 64 trend types; the 64 x 64 bit classifier actually classifies and codes 262820 combinations of 13 minutes in a day into 4096 trend types.
By using the 64-bit map code coding technology, 4096 complex stock price trend types are expanded and defined, so that the stock price change is subjected to fine analysis. 6k lines are taken, so that sectional analysis and combined analysis are facilitated, the technical characteristics of the stock price trend in a transaction period are analyzed by a single k line, and the trepanning characteristics of the stock price trend in a transaction period are analyzed by 6k lines. The four-color combined graph is convenient for a user to observe, analyze and distinguish to judge the subtle changes and differences of various trend types. The letter code is convenient for the computer to store and identify, and is also convenient for the user to search and read. The Chinese characters can be named, so that the user can conveniently perform inductive memory and judgment and identification, and the future trend of various trend types can be predicted. The Chinese phonetic alphabet can be used for compiling the names of trend types into phonetic simple codes, so that a user can search, retrieve and select a transaction model, and various strategy schemes can be quickly compiled.
The output end of the time-sharing k line combined graph code analysis module 3 is connected with the input end of the time-sharing transaction model determination module 4, and the output end of the time-sharing transaction model determination module 4 is connected with the input end of the transaction model screening stock module 10; the time-sharing k line combination pattern code analysis module 3 selects the trend type of the current day time-sharing k line combination with large rising and falling probability of the next transaction day from the output data of the 64-bit k line combination classification coding module 2, and the time-sharing transaction model determination module 4 determines the trend type of the current day time-sharing k line combination with large rising and falling probability of the next transaction day as a real-time transaction model;
The output end of the information acquisition module 5 is connected with the input end of the 64-bit k-wire combined image code classifying and step-by-step billing type data storage module 6, the output end of the 64-bit k-wire combined image code classifying and step-by-step billing type data storage module 6 is connected with the input end of the multi-period k-wire combined image code index module 7, the information acquisition module 5 acquires basic face information which has influence on stock price change in real time, including news, bulletins, information, researches and the like, defines the properties of interest and profits, places large disc information, industry information, regional information and individual stock information into the property according to the causality and relativity of the influence of the stock price, marks of red and green are used for distinguishing the profits and the property of the interest, the content of the information is used for reflecting the inherent relation of stock price change of information face factors, the objective relation of the analysis prediction is increased, and correct information support is provided for selection of a transaction model.
The 64-bit k-line combined picture code classification step-by-step account book data storage module 6 sets 64 classification daily accounts and 4096 detail classification daily accounts according to a 64-bit k-line combined coding principle, sets transaction daily accounts of each stock correspondingly, and records basic face information, transaction amount, rising and falling values and rising and falling amplitudes into corresponding daily accounts after each transaction day is finished; after the end of each week, the transaction amount, the rising and falling value and the rising and falling amplitude are transferred to the corresponding week classification account; after the end of each month, the transaction amount, the rise and fall value and the rise and fall value are recorded into the corresponding month classification account. The journal abstract column is two columns, which records important basic surface information and transaction data information, and provides support for large data analysis of artificial intelligence by using a classified and step-by-step accounting data storage mode.
The output end of the multi-period k-line combined picture code index module 7 is connected with the input end of the multi-period k-line combined picture code analysis module 8, the multi-period k-line combined picture code index module 7 uses k-line combined picture codes to compile stock price type tracking indexes, rising frequencies and profitability of a previous month k, a previous week k and a previous day k-line combined picture codes adjacent to a next month, a next week and a next day are arranged from high to low, rising and falling frequencies and profitability trends of subdivision trend types are more visual, and quantifiable rising and falling probability and profitability data are provided for analysis decisions of users; the multi-period k-line combined graph code analysis module 8 selects a transaction model of a month, week and day k-line combined graph code of a stock price trend type which is most easy to rise or fall in a later transaction period through index tracking data and basic surface information data;
The output end of the multi-period k-line combined graph code analysis module 8 is connected with the input end of the multi-period transaction model determination module 9, and the output end of the multi-period transaction model determination module 9 is connected with the input end of the transaction model screening stock module 10; the multi-period trading model determining module 9 determines a quantitative trading model for buying or selling according to the tracking index and the back-measurement analysis of the trend type, the trend type defined by the month, week and day k line combined graphic codes with high rising probability of the next period and high yield rate or high defect rate of the next period is selected from the trend types, and the trading model screening stock module 10 filters the trading models input by the multi-period trading model determining module and the time-sharing trading module step by step and screens the trading models into trading target stocks in real time.
As a specific design, the four-color k-ray diagram is formed, namely: the K line of small expansion is represented by red, the K line of large expansion is represented by purple, the K line of small drop is represented by green, the K line of large drop is represented by blue, the 64-bit K line combination codes are expanded into 4096 (64 x 64) K line combination codes, 4096 trend types are defined, and accordingly fine classification of stock price trend types is achieved, and support is provided for an visualized, identifiable and interpretable big data technology analysis system.
As a specific design, the transaction data acquisition module acquires transaction data of a stock, including: the stock exchange period, the time length, the expansion and drop amplitude, the hand change rate, the large bill transaction amount, the expansion and drop stop board, the institution transaction data published by the dragon and tiger list and various index data.
As a specific design, the abstract column of the journal is two columns, the basic surface information and transaction data information are recorded respectively, and the classified step-by-step accounting data are used for storing the artificial intelligence to provide support for large data analysis.
As a specific design, the transaction platform sub-machine 12 includes a transaction model screening stock selection module, a basic surface factor screening stock selection module, a stock pool module of a transaction target, a transaction instruction setting module, a transaction instruction newspaper module, a transaction instruction activating module and a transaction completion confirmation module, the platform server 11 controls the transaction platform sub-machine 12 for buying and selling transactions, a man-machine interaction platform is provided for users, the users set the transaction model, the technical index, the information factor and the basic surface factor through the transaction platform sub-machine 12 to form a transaction strategy scheme, the transaction platform sub-machine 12 automatically executes, and screens the stocks to perform full-automatic transactions.
The user can set the functions of the transaction platform sub-machine at the platform server so as to carry out efficient investment. The trade model screening stock selection module is characterized in that a user sets the selected trade model according to the trade idea and style of the user in a function column, the k line combination graphic codes of month, week, day and time sharing period are arranged and selected, the platform automatically performs real-time screening, and the trading target stock is provided for full-automatic trade. The basic surface factor screening stock selection module is characterized in that a user sets the selected basic surface factors according to the trading theory and style of the user in a function column, and after the information factors, the market rate, the dynamic market rate, the net market rate, the total stock and the total market value and the circulation market value are selected, the platform automatically performs real-time screening, so that trading target stocks are provided for full-automatic trading. The stock pool module of the trade target can observe the real-time dynamic of the stock of the trade target in the stock pool of the user, can use the manual wind control function to add and remove the stock in the stock pool, finally enter the stock of the trade target stock pool, and become the full-automatic trade target stock. The transaction instruction setting module is used for setting the number and the amount of transaction funds and stocks and a rule mode for activating the buying and selling instruction in the function field of the transaction instruction setting module by a user, and providing a preset instruction for full-automatic transaction. And the transaction instruction newspaper module is used for automatically matching the number, the quantity and the amount of stocks according to a preset transaction rule by the platform according to the buying and selling directions, and waiting for activation to finish the transaction. The trading order activating module is used for automatically executing trading orders according to a preset trading rule mode by the platform according to trading directions, such as index trading signal activating orders, big order trading activating orders and the like, and when target stocks meet preset algorithm conditions, the platform automatically activates the trading orders to confirm trading. If the target stock does not meet the condition meeting the preset algorithm, the platform does not activate the transaction instruction. The transaction confirmation completion module is used for conveying a transaction result to the platform server after the transaction is completed, and after the transaction is completed, the transaction platform sub-opportunity stores the purchased stocks into a selling column of a stock pool of a transaction target to wait for full-automatic selling transaction.
The user screens all stocks on the trading platform sub-machine by using the month k combined graph codes, and screens out the stocks of long-term trend types meeting the set requirements; the stocks of the selected long-term trend type are screened by using the week k combined graph codes, and the stocks of the medium-term trend type meeting the set requirements are screened; screening the selected stocks of the medium-term trend type by using a daily k combined graph code, and screening the stocks of the short-term trend type meeting the set requirements; screening the selected short-term trend type stocks by using the time-sharing k combined graph codes, and screening out the ultra-short-term trend type stocks meeting the set requirements; and finally, activating the transaction by using a set transaction instruction, and only trading the stocks with large single trading amount in the stock pool of the transaction target, thereby improving the accuracy and reliability of capturing the opportunity of individual transaction.
According to the full-automatic transaction flow, the platform server of the multi-period decision AI stock-frying robot platform constructs a neural network algorithm by using a multi-period mathematical model matrix, adopts a 64-bit k-line combination chart classification coding and step-by-step billing type data storage technology, uses big data to decompose and classify nuances of stock price trend characteristics, writes code names, reflects trend forms of various types of stock price trend in a tracking index mode, and extracts an optimal quantitative transaction model from the trend forms for screening stocks to form a stock market trend type identification system.
At 31 days of 1.2019, a trading model screening stock selection module of the trading platform machine sets a screening model of a buying trading strategy: six months coded by the month k combination are the types of trends of 'yin, yin' and 'continuous and greatly-falling' trends; six-week k lines coded by week k are "yin, yang, yin" and the type of trend is "rebound drop" trend; the six-day k line coded by the day k combination is 'positive, negative, positive and negative', and the trend type is 'anti-falling' trend; six time-sharing k lines of time-sharing k combination codes are selected to be 'yin, yang, yin and yang', the trend type is 'falling rebound' trend, and then the buying rule is selected to be that the large buying list with the time-sharing k falling amplitude exceeding-2% activates buying.
The basic surface factor screening stock selection module sets a basic surface factor of a purchase transaction: the information factors are more than 3 interest factors within 6 months, the market rate is less than 12 times, the net market rate is less than 2 times, and the circulation market value is less than 100 hundred million yuan. After the setting, the stocks meeting the conditions are few, and all the stocks meeting the conditions can automatically enter a stock pool of a trade target to execute full-automatic trade.
In the stock pool, 000636 Fenghuagao is selected, the trend of the large plate on the same day is stable rebound of the enterprise after the large plate drops greatly, and the trend of the 000636 Fenghuagao is of the type:
The combination of the rising and falling of month k of the long-term strategy is-60.84%, the combination of the rising and falling of month k of the long-term strategy is-16.18%, 8.12%, 8.51%, 14.83%, 18.36% and 4.84% of the falling in turn, and the morphological trend type is 'continuous falling';
The combination of the rising and falling amplitudes of week k of the medium-term strategy is-4.85%, the rising and falling amplitudes of week k are-3.07%, 5.21%, 5.22%, 3.62%, 2.09% and 10.68% in turn, and the morphological trend type is rebound and falling;
The combination of the daily k combination of the short-term strategy is-12.33%, the daily k combination of the short-term strategy is 1.72%, 1.1%, 0.85%, 9.91%, 0.48% and 2.67% of the daily k combination of the short-term strategy, and the morphological trend type is 'resistant to falling';
The total of the daily time-sharing k (40 minutes) combined expansion and falling amplitude of the ultra-short term strategy is-2.67%, the time-sharing is sequentially-2%, 0.57%, 1.05%, 0.37%, 0.2% and 0.38% of the time-sharing, and the morphological trend type is 'falling rebound'.
Basic surface information data of day 000636 Fenghuagao is: from the highest price 24.81 yuan of 18 days of 7 months of 2018 to the lowest price 9.86 yuan of 31 days of 1 month of 2019, the accumulated drop width is-60 percent, and the continuous drop is nearly six half months; within 6 months, the product supply of the company is insufficient, the price of the product is continuously increased, the performance is greatly increased, the industry yield is insufficient, and the like, and the interest factors are more than 3; the urban ratio is about 11 times and less than 12 times; the net market rate is about 1.7 times and less than 2 times; the market value is about 92 billions and less than 100 billions.
The technical face data and basic face information data of 000636 Fenghuagao all meet the set trend type of multi-period screening and the estimation standard of basic face factor screening, the stock pool entering the trade target is automatically screened by the platform to become the stock of the trade target, and the buying instruction is automatically executed in the first time-sharing (9:30-10:10) period.
The commodity falls greatly after being opened in 1 month 31 of 2019, falls by-6% in 5 minutes of diving in the first time-sharing period, falls to 9.86 yuan, triggers a buying command to carry out dynamic newspaper, receives and initiatively purchases a super large buying command in 35 minutes at 9 points, rises to 9.99 yuan from 9.86 yuan, reaches 2.17 ten thousand hands in 1 minute, reaches 2166 ten thousand yuan in the trading amount, and activates a stir-frying robot to buy the trading command to confirm buying, and the buying price is 9.99 yuan. The price of the stock is 10.40 yuan on the same day, the highest price is 10.40 yuan, the lowest price is 9.86 yuan, the receiving price is 10.22 yuan, and the drop is-2.67%. The total amount of the assembly is 37.27 ten thousand hands, the total amount of the assembly is 38036 ten thousand yuan, the hand replacement rate is 4.26%, and the bottom amount is put. Thereafter, the strand spreads out to rebound greatly and the bottom morphology is established. After the stock is purchased and traded, the stock automatically enters the selling column of the target stock pool to wait for the trading platform to give out a selling trading instruction.
Since stock market trading operation is the most important link for buying stock, the safety margin and the statistical probability of buying stock price are considered mainly, the high efficiency of fund operation and the maximization of income are achieved, the selection condition requirement is strict, the basic surface factors and the multi-period trading model are required to be used for screening together, and the long-term factors and the medium-term factors of the statistical rule of stock price fluctuation are more emphasized. However, in the link of selling stocks, the statistical rule short-term and ultra-short-term factors of stock price fluctuation are considered, and the randomness and uncertainty of the stock price fluctuation are the important matters of operation. Therefore, the medium, short and ultrashort are mainly selected in the process of selling the trading model, the trading period is set to be 1 month and 2 weeks, and therefore after the trading model of the trading platform sub-machine receives the trade data in 1 month and 31 days in 2019, the trading model screening stock selection module of the trading platform sub-machine sets a screening model of selling the trading strategy: month k combination codes are of a trend type of 'yin, yin and yang' for six months, and are of a morphological trend type of 'falling-stopping and falling' -trend; the combination codes of the week k are the morphological trend types of six weeks k lines of positive, negative, positive and positive, and the trend type of the six weeks k lines of positive, negative, positive and positive is the trend type of the vibration rising; the combination codes of the day k are the trend types of 'positive, negative, positive and positive' of the six-day k line, and the trend type of 'relay rising'; the combination codes of time sharing k are selected to be six trend types of time sharing k lines of 'positive, negative, positive and positive', and form trend types of 'shock rising', and then the buying rule is selected to activate and sell the large sales order with the time sharing positive k fluctuation exceeding 2%. After the transaction model and the transaction rule are selected in this way, the transaction model and the transaction rule are fully matched with the preset trend type of multi-period screening by the day 19 of 2 months in 2019 on the market column of the stock pool of the transaction target waiting for the transaction of 000636 Fenghai, and the sales order is automatically executed in the sixth time sharing (14:20-15:00) period.
The current day of large plate trend is continuous rebound trend after successful bottom detection, and the current day of 000636 bloom is of the high-tech trend type:
Month k combination of long-term strategy comprises the steps of falling-8.12%, falling-8.51%, falling-14.83%, falling-18.36%, falling-4.84% and rising 44.32% in sequence, and the morphological trend type is 'anti-falling rebound';
The combination of week k of the mid-term strategy is divided into 5.22% of rising, 3.62% of falling, 2.09% of rising, 10.68% of falling, 20.10% of rising and 17.29% of rising in turn, and the morphological trend type is 'shake rising';
The combination of the days k of the short-term strategy is that the day is 9.10 percent, 7.73 percent, 2.53 percent, 0.65 percent, 10 percent and 4.48 percent of the rise, and the form trend type is 'relay rise';
The time-sharing k (40 minutes) combination in the day of the ultra-short term strategy sequentially comprises 1.33% of swelling, 3.50% of swelling, 1.83% of falling, 0.21% of falling, 0.14% of swelling and 6.04% of swelling, and the trend type is 'shake rising'.
After 19 days of opening the dish in 2019, 2 months: in the first time-sharing (9:30-10:10) period, the fluctuation range is less than 2%, and the selling instruction is not triggered; in the second time-sharing (10:10-10:50) period, the fluctuation amplitude is more than 2%, the selling instruction is triggered to carry out dynamic newspaper, but the super large sales order throwing disc and the active large sales order do not appear, so that the transaction is not activated; the selling instruction is not triggered when the third time-sharing (10:50-11:30) period is a drop, the fourth time-sharing (13:00-13:40) period is a slight drop, and the fifth time-sharing (13:40-14:20) period is rise slightly; the method comprises the steps of suddenly pulling up to 14.22 yuan in 14 minutes at a time point of a sixth time sharing (14:20-15:00) period, triggering a selling instruction to dynamically report a bill in a time-sharing rise of 2.22%, impacting a swelling stop board in 48 minutes at the time point of 14, enabling a 1-minute trading volume of 6.69 ten thousand hands, enabling a trading amount of 9899 ten thousand yuan, enabling a super large selling bill throwing board and an active large selling bill to appear in 14.87 yuan of the swelling stop board, and activating the selling transaction instruction to confirm trading, wherein the selling price is 14.87 yuan. The price of the stock is opened by 13.59 yuan on the same day, the highest price is 14.87 yuan, the lowest price is 13.39 yuan, the closing price is 14.75 yuan, and the rise amplitude is 9.10%. The total amount of the assembly is 139.25 ten thousand hands, the total amount of the assembly is 195583 ten thousand yuan, and the hand replacement rate is 15.56%. Thereafter, the strand ends in a largely strongly rising form and begins in a shock rising form. This round of operation, from 9.99 cells to 14.87 cells, resulted in 48.85% profits for 9 transaction days.
000636 Tendency of the high family of wind bloom: from the highest price of 24.81 yuan in 7.18 in 2018 to the lowest price of 9.86 yuan in 31 in 1.31 in 2019, the accumulated drop width is-60%, the continuous drop is nearly six half months, and the rebound is unfolded after the bottom detection is successful, so that the long-term bear market drop market is ended. From 31 days 1 month to 25 days 2 months 2019, from 9.86 yuan of lowest price to 15.60 yuan of highest price, a rise of 58.21% is completed by 13 transaction days; setting back to the lowest price of 13.59 yuan for 3 months and 1 day for 25 months, adjusting for 5 days, vibrating to rise to 3 days for 4 months and 17.69 yuan for the highest price, and rising for 30.17% by 24 transaction days; from 9.86 yuan of lowest price to 17.69 yuan of highest price, 40 transaction days are used in total, the accumulated rise is 79.41%, and the rising market of the middle-grade rebound of the round is ended.
The full-automatic transaction platform of the AI stranding robot provided by the invention takes 6 k-wire combinations as codes to form classification trend types, and is used in aspects. The user only needs to manually select and set the transaction model, the basic surface factors and the transaction instructions, then a high-winning quantized transaction model decision scheme can be formed, and the stranding robot can automatically execute the scheme. The specific method comprises the following steps: the user selects and combines the long-term month k combination chart codes, the mid-term week k combination chart codes, the short-term day k combination chart codes and the ultra-short-term day time-sharing k combination chart codes in the screening model, and the technical investment strategy meeting the investment requirement of the user can be generated by referring to the trend direction of the classification tracking index and the counted rising probability and the counted yield of the last month, the last week and the last day; and then the basic surface factors are selected and set to generate a basic surface investment strategy meeting the investment requirements of the user.
Through the settings, a plurality of quantization models of high-winning comprehensive trading strategies can be generated, after the quantization models are determined, the stock frying robot automatically executes the stock frying, full market searching and screening are carried out, stocks meeting the set conditions are put into a stock pool of a trading target, and finally the platform automatically activates trading according to the set trading instructions. The transaction platform sub-machine can capture rising opportunities at the rising point of an individual strand at a high speed, so that transaction advantages which cannot be obtained by manual analysis are obtained, and ultrahigh investment benefits are realized.
The invention provides a stir-frying robot, which is characterized in that a logic deriving system with strict flow is formed by connecting all functional modules, a k-line combined graph with multiple periods is encoded by applying an artificial intelligence machine encoding technology to form a detailed and various trend type, the current period trend type with high rising probability and high yield of the later transaction period is generated by computer identification and analysis, and a high-winning comprehensive investment strategy is formed after the user manually selects the current period trend type, so that the AI stir-frying robot transaction platform with the characteristics of automatic identification, automatic induction and automatic excitation is finally realized.
The user uses a high-winning policy model which is selected and combined in a few minutes and a few hours, the system can search and screen target stocks meeting the conditions in real time in the whole market, and a high-winning target stock pool can be generated in a few seconds, which is the workload that manual analysis cannot be completed in days, weeks or even months, and the transaction efficiency is greatly improved. The robot randomly selects stocks to wait for transaction and randomly activates and confirms the transaction, is a decision tool conforming to the game theory, replaces the subjective decision of individuals by utilizing the algorithm advantage of artificial intelligence, reduces investment errors caused by personal emotion fluctuation and cognitive deviation, and thus obtains investment advantages which cannot be obtained by manual analysis and manual operation.
While the invention has been described and illustrated in detail in the foregoing description with reference to specific embodiments thereof, it should be noted that various equivalent changes and modifications could be made to the above described embodiments without departing from the spirit of the invention as defined by the appended claims.
Claims (6)
1. A multi-cycle decision AI (advanced technology attachment) stranding robot platform is characterized in that: the full-automatic trading platform of the multi-period decision AI (automatic identification) stock-frying robot adopts a 64-bit k-line combined graph classification coding and step-by-step billing data storage technology, a big data statistics method is used for decomposing and classifying nuances of stock price trend characteristics, code names are written, trend forms of various trend types and stock prices are intuitively reflected in a tracking index mode, an optimal quantitative trading model is extracted from the trend forms for screening stocks, the optimal stocks are imported into a stock pool of a trading target after screening and filtering, and finally the full-automatic trading of the AI stock-frying robot is realized through trading instruction activation;
The full-automatic trading platform of the multi-period decision AI stranding robot comprises a platform server, a trading platform sub-machine, wherein a plurality of trading platform sub-machines are connected with the platform server, a trading data acquisition module (1), a 64-bit k line combination classification coding module (2), a time-sharing k line combination graph code analysis module (3), a time-sharing trading model determination module (4), an information acquisition module (5), a 64-bit k line combination graph code classification step-by-step billing type data storage module (6), a multi-period k line combination graph code index module (7), a multi-period k line combination graph code analysis module (8), a multi-period trading model determination module (9) and a trading model screening stock module (10) are arranged in the platform server;
The output end of the transaction data acquisition module (1) is connected with the input end of the 64-bit k-line combination classification coding module (2), transaction data of the abnormal stocks are acquired, abstract data are provided for billing storage, and correct data support is provided for selection of a transaction model; the output end of the 64-bit k-line combination classification coding module (2) is connected with the input end of the time-sharing k-line combination graph code analysis module (3), and the 64-bit k-line combination classification coding module (2) codes 6 k-line combination graphs into 64 k-line combination codes according to the rising and falling characteristics according to coding rules, so as to define 64 trend types; according to the rising and falling amplitude characteristics of each k line, a four-color k line graph is compiled, 64-bit k line combination codes are expanded into 4096 k line combination codes, 4096 trend types are defined, and therefore fine classification of stock price trend types is achieved;
The output end of the time-sharing k line combined graph code analysis module (3) is connected with the input end of the time-sharing transaction model determination module (4), and the output end of the time-sharing transaction model determination module (4) is connected with the input end of the transaction model screening stock module (10); the time-sharing k line combination graph code analysis module (3) selects the trend type of the current day time-sharing k line combination with high rising and falling probability of the next transaction day from the output data of the 64-bit k line combination classification coding module (2), and the time-sharing transaction model determination module (4) determines the trend type of the current day time-sharing k line combination with high rising and falling probability of the next transaction day as a real-time transaction model;
The output end of the information acquisition module (5) is connected with the input end of the 64-bit k-line combined image code classified step-by-step billing type data storage module (6), the output end of the 64-bit k-line combined image code classified step-by-step billing type data storage module (6) is connected with the input end of the multi-period k-line combined image code index module (7), the information acquisition module (5) acquires basic surface information influencing the change of the stock price in real time, defines the properties of emptiness and interest, and is respectively placed into large disc information, industry information, regional information and individual stock information according to the causal relations of the influence of the stock price, the content of the marked by marking symbols with different colors is distinguished by the properties of the interest and the interest, the content of the marking column is used for reflecting the inherent relation of the change of the stock price, the objective of analysis prediction is increased, and correct information support is provided for the selection of a transaction model;
The 64-bit k-line combined graph code classification step-by-step billing data storage module (6) sets 64 classification daily accounts and 4096 detail classification daily accounts according to a 64-bit k-line combined coding principle, sets transaction daily accounts of each stock correspondingly, and records basic face information, transaction amount, expansion and fall values and expansion and fall amplitudes into corresponding daily accounts after each transaction day is finished; after each week of transaction day is finished, the transaction amount, the rising and falling value and the rising and falling amplitude are transferred into corresponding week classification accounts; after each month of transaction day is finished, the transaction amount, the rising and falling value and the rising and falling amplitude are transferred into corresponding month classification accounts;
the output end of the multi-period k-line combined picture code index module (7) is connected with the input end of the multi-period k-line combined picture code analysis module (8), the multi-period k-line combined picture code index module (7) compiles a stock price type tracking index by using k-line combined picture codes, the rising frequency and the income rate of the previous month k, the previous week k and the next month, the next week and the next day adjacent to each other of the previous day k-line combined picture codes are arranged from high to low, so that the rising and falling frequency and the income rate trend of subdivision trend types are more visual, and quantifiable rising and falling probability and income rate data are provided for analysis decisions of users; the multi-period k-line combined graph code analysis module (8) selects a transaction model of a month, week and day k-line combined graph code of a stock price trend type which is most easy to rise or fall in the later period through index tracking data and basic surface information data;
The output end of the multi-period k-line combined graphic code analysis module (8) is connected with the input end of the multi-period transaction model determination module (9), and the output end of the multi-period transaction model determination module (9) is connected with the input end of the transaction model screening stock module (10); the multi-period trading model determining module (9) determines a quantitative trading model for buying or selling according to the trend types defined by the current month, current week and current day k line combined graphic codes, wherein the rising probability of the next period is high, the yield is high, or the falling probability of the next period is high, the defect rate is high, and the trading model screening stock module (10) filters the trading models input by the multi-period trading model determining module and the time-sharing trading module step by step and screens the trading models as trading targets in real time.
2. The multi-cycle decision AI stock robot platform of claim 1, wherein: the four-color k line graph is compiled, namely: the small-rise K-line, the purple-color K-line, the green-color K-line, and the blue-color K-line are indicated by red, green, and blue, respectively.
3. The multi-cycle decision AI stock robot platform of claim 1, wherein: the transaction data acquisition module acquires transaction data of the stock exchange, and comprises: the stock exchange period, the time length, the expansion and fall amplitude, the hand change rate, the large bill transaction amount, the expansion and fall stop board, the institution transaction data published by the dragon and tiger list and various index data.
4. The multi-cycle decision AI stock robot platform of claim 1, wherein: the abstract columns of the journal are two columns, the two columns respectively record basic surface information and transaction data information, and classified step-by-step accounting data are stored as big data for artificial intelligent analysis.
5. The multi-cycle decision AI stock robot platform of claim 1, wherein: the transaction platform sub-machine (12) comprises a transaction model screening stock selection module, a basic surface factor screening stock selection module, a transaction target stock pool module, a transaction instruction setting module, a transaction instruction newspaper module, a transaction instruction activating module and a transaction completion confirmation module;
The transaction model screening stock selection module receives data of a platform server, compiles and selects k line combined graph codes of month, week, day and time sharing period, and the transaction model is set by a user according to the transaction idea and style of the user;
The basic surface factor screening stock selection module receives transaction model information of the transaction model screening stock selection module, a user sets the selected basic surface factors according to own transaction concepts and styles, and information factors, market rates, dynamic market rates, market net rates, total stakes, total market values and circulation market values are selected;
The stock pool module of the trade target receives the selected data of the basic surface factor screening stock selection module, and performs real-time screening to form the stock pool of the trade target;
The trade command setting module is connected with the trade target stock pool module, selects stocks from the trade target stock pool, sets trade funds, the number and the amount of the stocks, activates the trading command and provides a preset command for full-automatic trade;
The transaction instruction newspaper module receives the data information of the transaction instruction setting module, and automatically matches the number, the quantity and the amount of stocks according to a preset transaction rule and a buying and selling direction to form newspaper information;
The transaction instruction activation module receives the newspaper information of the transaction instruction newspaper module, sends the newspaper information into the platform server, the platform server automatically executes the transaction instruction according to the buying and selling directions according to the preset transaction rule, when a certain target stock in the newspaper information meets the preset algorithm condition, the platform server sends out the transaction instruction to the transaction instruction activation module, and when a certain target stock in the newspaper information does not meet the preset algorithm condition, the platform server sends out the transaction instruction to the transaction instruction activation module;
The confirmation completion transaction module receives an activation transaction instruction of the transaction instruction activation module, purchases or sells a certain target stock in newspaper information meeting the buying or selling conditions, after the transaction is completed, a transaction result is transmitted to the platform server, after the transaction is completed, the target stock pool module of the transaction stores the purchased stock into a selling column of a target stock pool of the transaction, and the transaction waits for full-automatic selling transaction; for the stocks that have completed the sell trade, the trade target stock pool module deletes them from the sell column of the trade target stock pool; the platform server (11) controls the trading of the trading platform sub-machine (12), a user sets a trading model, technical indexes, information factors and basic factors through the trading platform sub-machine (12) to form a trading strategy scheme, and the trading platform sub-machine (12) automatically executes, screens stocks and carries out full-automatic trading.
6. The multi-cycle decision AI stock robot platform of claim 5, wherein: in the stock pool of the trade target stock pool module, a user observes the real-time dynamics of the trade target stock and allows the stock to be added and removed, and finally the stock of the trade target stock pool is formed and becomes the full-automatic trade target stock.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110223333.0A CN112734577B (en) | 2021-03-01 | 2021-03-01 | Multi-period decision AI (advanced technology attachment) strand-frying robot platform |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110223333.0A CN112734577B (en) | 2021-03-01 | 2021-03-01 | Multi-period decision AI (advanced technology attachment) strand-frying robot platform |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112734577A CN112734577A (en) | 2021-04-30 |
| CN112734577B true CN112734577B (en) | 2024-10-18 |
Family
ID=75595568
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110223333.0A Active CN112734577B (en) | 2021-03-01 | 2021-03-01 | Multi-period decision AI (advanced technology attachment) strand-frying robot platform |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112734577B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2024098372A1 (en) * | 2022-11-11 | 2024-05-16 | Lora Research Limited | System and method for selecting automated financial product trading module |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106960388A (en) * | 2017-03-01 | 2017-07-18 | 中钞信用卡产业发展有限公司北京智能卡技术研究院 | The method and apparatus of the digital asset circulation of transregional piece of chain |
| CN109615528A (en) * | 2019-02-21 | 2019-04-12 | 上海盈首信息科技有限公司 | A kind of AI speculation in stocks full-automatic transaction platform of robot |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104572886B (en) * | 2014-12-23 | 2018-01-30 | 浙江大学 | The financial time series similarity query method represented based on K line charts |
| CN109685656A (en) * | 2018-12-21 | 2019-04-26 | 天津知柿信息科技有限公司 | A kind of stock market's tendency intelligent Forecasting and system |
| CN111815447A (en) * | 2020-07-06 | 2020-10-23 | 上海汇正财经顾问有限公司 | Stock intelligent recommendation system and method based on backtesting data and electronic terminal |
| CN112102086A (en) * | 2020-09-17 | 2020-12-18 | 上海盈首信息科技有限公司 | Design method and system of full-automatic trading model of AI stock-frying robot |
-
2021
- 2021-03-01 CN CN202110223333.0A patent/CN112734577B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106960388A (en) * | 2017-03-01 | 2017-07-18 | 中钞信用卡产业发展有限公司北京智能卡技术研究院 | The method and apparatus of the digital asset circulation of transregional piece of chain |
| CN109615528A (en) * | 2019-02-21 | 2019-04-12 | 上海盈首信息科技有限公司 | A kind of AI speculation in stocks full-automatic transaction platform of robot |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112734577A (en) | 2021-04-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Guo et al. | Does digital transformation improve the firm’s performance? From the perspective of digitalization paradox and managerial myopia | |
| Day et al. | AI robo-advisor with big data analytics for financial services | |
| Zarutska et al. | Structural-functional analysis as innovation in public governance (case of of banking supervision) | |
| CN106022522A (en) | Method and system for predicting stocks based on big data published by internet | |
| CN111291925A (en) | Financial market prediction and decision-making system and method based on artificial intelligence | |
| CN110689437A (en) | Communication construction project financial risk prediction method based on random forest | |
| Terdpaopong et al. | Modelling SME credit risk: Thai empirical evidence | |
| CN114943458A (en) | Enterprise ESG (electronic service guide) rating method based on weight distribution model | |
| CN112734577B (en) | Multi-period decision AI (advanced technology attachment) strand-frying robot platform | |
| Chatigny et al. | Asset pricing with attention guided deep learning | |
| CN111951093A (en) | Personal credit score scoring method | |
| Olson et al. | The linkages between dividends and earnings | |
| Davis | The strategic divestment decision | |
| CN113837803A (en) | Second-hand mobile phone recycling price prediction algorithm based on multi-model fusion | |
| Bhargava et al. | Quantifying narratives and their impact on financial markets | |
| Bakhshi et al. | Developing a hybrid approach to credit priority based on accounting variables (using analytical network process (ANP) and multi-criteria decision-making) | |
| Yun et al. | Stock Prediction via Machine Learning and Factor Analysis | |
| Hsu et al. | Classification methods of credit rating-A comparative analysis on SVM, MDA and RST | |
| Cao | Financial risk evaluation of listed companies in software and information technology service industry in China-Take M Company as an example | |
| Alimoradi Gaghdari et al. | Providing a Model for Optimal Capital Structure in Gas Refining Companies and Determining Parameters Using the AHP Method | |
| Su et al. | Spread Movement Prediction for Pairs Trading with High-Frequency Limit Order Data | |
| Adakh et al. | Comparison of Some Data Mining Models in Forecast of Performance of Banks Accepted in Tehran Stock Exchange Market | |
| Sadatrasoul et al. | Investigating Revenue Smoothing Thresholds That Affect Bank Credit Scoring Models: An Iranian Bank Case Study | |
| Ma et al. | Credit Risk Assessment of Chinese Listed Companies Based on SVM Improved by Shuffled Frog Leaping Algorithm | |
| Yacobus et al. | Analysis of The Effect Of Financial Distress Conditions On Activity/Efficiency Ratios In Manufacturing Companies In Indonesian Stock Exchange During The Covid Period |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |