CN111815447A - Stock intelligent recommendation system and method based on backtesting data and electronic terminal - Google Patents
Stock intelligent recommendation system and method based on backtesting data and electronic terminal Download PDFInfo
- Publication number
- CN111815447A CN111815447A CN202010642346.7A CN202010642346A CN111815447A CN 111815447 A CN111815447 A CN 111815447A CN 202010642346 A CN202010642346 A CN 202010642346A CN 111815447 A CN111815447 A CN 111815447A
- Authority
- CN
- China
- Prior art keywords
- stock
- retest
- model
- strategy
- time
- 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
Images
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
Landscapes
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Engineering & Computer Science (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Technology Law (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The application discloses a stock intelligent recommendation system, method and electronic terminal based on backtesting data, the system includes: the method comprises the steps of screening a model, sequencing the model, performing a retesting model, and performing historical retesting on second stock data according to a configurable first retesting strategy to obtain a first historical retesting result; evaluating the first historical retest result according to at least one evaluation index, and correspondingly adjusting and updating the screening condition or/and the sequencing condition until the retest model is switched from the retest state to the working state when the first historical retest result output by the retest model meets a first expected condition; the recommendation model selects a third stock of ticket data meeting the real-time recommendation condition from the second stock of ticket data according to a configurable real-time push strategy to carry out push output; and the test-back model tests back the third stock of ticket data according to a configurable second test-back strategy. The method and the system effectively recommend more objective and high-accuracy stocks to the user, and simultaneously set a proper profit-loss ratio so as to realize the maximization of the relative profitability.
Description
Technical Field
The application relates to the financial field, in particular to the technical field of stock data analysis, and specifically relates to a stock intelligent recommendation system and method based on backtesting data and an electronic terminal.
Background
With the rapid development of market economy in China, more and more investors aim at stocks, and the stock investment becomes an important mode in personal financing. Stocks can be circulated and transferred in the market, and the stocks have the characteristics of high risk and high return, so that tens of millions of investors are involved from the generation day, and the investors pay attention to the fluctuation and development trend of the stock price all the time, which is always a hot problem in the research of the economic field.
The large income and small risk are the targets pursued by security investors. Two types of investment analysis methods, basic analysis methods and technical analysis methods, are commonly used to achieve this goal. The basic analysis method analyzes the basic factors influencing the supply and demand relationship of the stock market, determines the real value of the stock, judges the trend of the stock market and provides the basis for investors to select the stock. The technical analysis is a method for analyzing completely according to the stock market quotation change, and it can judge the future change trend of whole stock market or individual stock price by analyzing historical data, and discuss the possible turn of investment behavior in stock market, and provide the signal for buying and selling stocks for investors.
Technical analysis methods, such as a moving average line method, a point graph method, a K line graph method and the like, have the characteristics of simplicity, clarity and easy mastering, but because stock market operation is a huge nonlinear system, stock price trend is influenced by various factors such as politics, economy, psychology and the like, analysis results of different people using the technical analysis methods have obvious differences, and therefore, the technical analysis application results of common investors in stock selection and risk avoidance are not ideal.
In the actual stock investment process, a user usually decides which stock to buy according to subjective judgment or recent stock consultation, or the investment manager subjectively recommends which stock to buy to the user according to some related information of the benchmark stocks, however, the uncertainty of stock selection based on human experience is large. Therefore, how to recommend more objective and highly accurate stocks for users becomes a technical problem which needs to be solved urgently.
Disclosure of Invention
In view of the above-mentioned shortcomings of the prior art, the present application aims to provide an intelligent stock recommendation system, method and electronic terminal based on backlog data, which are used for recommending more objective and accurate stocks for users.
To achieve the above and other related objects, the present application provides an intelligent stock recommendation system based on backlog data, comprising: the screening model is used for screening the current stock data according to a configurable screening strategy and outputting first stock data meeting screening conditions; the sorting model is used for sorting the first stock data meeting the screening conditions according to a configurable sorting strategy and outputting second stock data which meet the sorting conditions and are arranged in sequence; the backtesting model is used for performing historical backtesting on the second stock data according to a configurable first backtesting strategy to obtain a first historical backtesting result; evaluating the first historical retest result according to at least one evaluation index, and correspondingly adjusting and updating the screening condition or/and the sequencing condition until the retest model is switched from the retest state to the working state when the first historical retest result output by the retest model meets a first expected condition; the recommendation model selects third stock data meeting the real-time recommendation condition from the second stock data according to a configurable real-time push strategy to carry out push output after the retest model enters a working state; the backtesting model performs backtesting on the third stock of ticket data according to a configurable second backtesting strategy to obtain a second historical backtesting result; the retest model judges whether the second historical retest result meets a second expected condition, and if so, the retest model does not act; and if not, the retest model is switched to the retest state from the working state when the state switching condition is met.
In one embodiment, the first reward policy includes a plurality of combinations of reward time, a reward benchmark, a transaction cost, and an exclusion period; the evaluation indexes comprise various combinations of total income evaluation indexes, preset key evaluation indexes and positive correlation evaluation indexes; the preset key evaluation indexes comprise various combinations of sharp rate, maximum withdrawal rate, executable rate and income fluctuation rate.
In one embodiment, the first expected condition includes: a plurality of combinations of a win rate, a profit rate, a maximum fall and a maximum return depth for a preset time.
In one embodiment, the screening policy includes a configuration policy for one or more combinations of market value, time interval, and market value; the ranking strategy comprises a configuration strategy of deducting one or more combinations of net asset profitability, order, range and weight.
In an embodiment, the real-time push strategy in the recommendation model includes: calculating one or more push times for pushing the stock; and judging whether stocks meeting the pushing conditions in the real-time disk exist in second stock data of the day before the day, if so, pushing the stocks to the user terminal, and if not, selecting corresponding stocks according to the ordering of the stocks in the second stock data to push the stocks to the user terminal when the pushing time is up.
In one embodiment, whether to execute a real-time push strategy is controlled by a push probability function meeting push conditions in a real-time disk; the push probability function is a monotonous function of the real-time comprehensive expansion amplitude, the distance market receiving time and the push probability of each strand; suppose that the set latest recommended time is T0If the current time is T, then T0-T is the latest recommended time Δ T, and when Δ T is 0, selecting the corresponding stock from the second stock data and pushing the stock to the user terminal; and when the real-time comprehensive amplitude of each strand is smaller than the amplitude limit value, stopping executing the real-time pushing strategy.
In one embodiment, the recommendation model pushes the determined ending surplus price and ending loss price while recommending the stocks to the user terminal; and the return test model carries out return test on the determined stop loss price and the stop loss price, and determines whether the currently pushed stop loss price and stop loss price are suitable.
In one embodiment, one implementation of determining the reserve price and the reserve loss price includes: respectively calculating the total number of profit trades and the total number of loss trades of the stock trading history data of the stock according to the stock selection strategy; respectively calculating the average income of the stock profit transaction and the average income of the loss transaction according to the total number of the profit transaction and the total number of the loss transaction; determining a loss value near 1 standard deviation according to the distribution of the maximum loss of all stocks, determining the loss value as a reserve price, and reversely pushing a minimum reserve value required for realizing profit according to a profit-loss ratio, and determining the minimum reserve price.
In order to achieve the above and other related objects, the present application provides an intelligent stock recommendation method based on backlog data, comprising: and (3) a retesting step: screening the current stock data according to a configurable screening strategy, and outputting first stock data meeting screening conditions; sorting the first stock data meeting the screening condition according to a configurable sorting strategy, and outputting second stock data which meet the sorting condition and are sequentially arranged; performing historical backtesting on the second stock data according to a configurable first backtesting strategy to obtain a first historical backtesting result; evaluating the first historical retest result according to at least one evaluation index, and correspondingly adjusting and updating the screening condition or/and the sequencing condition until the retest model is switched from the retest state to the working state when the first historical retest result output by the retest model meets a first expected condition; a real-time pushing step: after the retest model enters a working state, selecting third stock of ticket data meeting real-time recommendation conditions from the second stock of ticket data according to a configurable real-time pushing strategy to carry out pushing output; carrying out retesting on the third stock of ticket data according to a configurable second retesting strategy to obtain a second historical retesting result; judging whether the second historical retest result meets a second expected condition, if so, not acting; and if not, the retest model is switched to the retest state from the working state when the state switching condition is met.
To achieve the above and other related objects, the present application provides an electronic terminal including: at least one display for displaying; at least one memory for storing a computer program; at least one processor, coupled to the display and the memory, is configured to run the computer program to implement the above-mentioned method for intelligent stock recommendation based on the measured data.
As described above, the stock intelligent recommendation system, method and electronic terminal based on the backtesting data have the following beneficial effects:
the stock selection strategy is subjected to historical backtesting in the application, the profit probability is high, the stocks are recommended in different periods through the configurable real-time pushing strategy, the condition that the stocks are bought too intensively and consistently by a user to influence stock trends is avoided, the pushed stocks are continuously backtested, more objective stocks with high accuracy are effectively recommended for the user, a proper profit-loss ratio is set, and accordingly the relative profit-yield maximization is realized.
Drawings
Fig. 1 is a schematic block diagram of an intelligent stock recommendation system based on survey data in an embodiment of the present application.
FIG. 2 is a schematic diagram of screening conditions exhibited by a screening model according to an embodiment of the present application.
FIG. 3 is a schematic diagram of the sorting conditions shown for the sorting model in an embodiment of the present application.
FIG. 4 is a diagram illustrating a regression model according to an embodiment of the present application.
Fig. 5 is a schematic diagram of a backtesting condition of a first backtesting strategy according to an embodiment of the present disclosure.
FIG. 6 is a schematic diagram of the data of the daily-based historical return survey of the on-line stub stock strategy according to the embodiment of the present application.
Fig. 7 is a flowchart illustrating a real-time push policy according to an embodiment of the present application.
Fig. 8 shows a graph of a push probability function for an embodiment of the application.
Fig. 9 is a schematic diagram of push time according to an embodiment of the present application.
Fig. 10 to 12 are schematic diagrams illustrating the test effect of the intelligent stock recommendation method based on backtesting data of the present application on the suspension of the engorgement and loss.
Fig. 13 is a schematic structural diagram of an electronic terminal according to an embodiment of the present application.
Detailed Description
The following description of the embodiments of the present application is provided by way of specific examples, and other advantages and effects of the present application will be readily apparent to those skilled in the art from the disclosure herein. The present application is capable of other and different embodiments and its several details are capable of modifications and/or changes in various respects, all without departing from the spirit of the present application. It should be noted that the embodiments and features of the embodiments in the present application may be combined with each other without conflict.
It should be noted that the drawings provided in the following embodiments are only for illustrating the basic idea of the present invention, and the components related to the present invention are only shown in the drawings rather than drawn according to the number, shape and size of the components in actual implementation, and the type, quantity and proportion of the components in actual implementation may be changed freely, and the layout of the components may be more complicated.
The embodiment aims to provide an intelligent stock recommendation system and method based on backtesting data and an electronic terminal, which are used for recommending more objective and high-accuracy stocks for users.
The principles and implementations of the system and method for intelligently recommending stocks based on measured data and the electronic terminal according to the present embodiment will be described in detail below, so that those skilled in the art can understand the system and method for intelligently recommending stocks based on measured data and the electronic terminal without creative work.
As shown in fig. 1, this embodiment provides an intelligent stock recommendation system based on measured data, which includes: the system comprises a screening model, a sequencing model, a retesting model and a recommending model.
The stock selection strategy in this embodiment is realized through a screening model and a ranking model, and the ranking model determines the stocks meeting the conditions through the screening model.
In this embodiment, as shown in fig. 2, the screening model screens the current stock data according to a configurable screening policy, and outputs the first stock data meeting the screening condition.
Wherein the screening conditions refer to market circulation value, time interval, market value and the like; the screening strategy comprises a configuration strategy for one or more combinations of the market value, the time interval and the market value.
In this embodiment, as shown in fig. 3, the sorting model sorts the first stock data meeting the screening condition according to a configurable sorting policy, and outputs the second sequentially arranged stock data meeting the sorting condition.
Wherein the sorting condition refers to deduction of non-net asset profitability, order, range, weight and the like, and the sorting strategy comprises configuration strategy of one or more combinations of deduction of non-net asset profitability, order, range and weight.
As can be seen from the above, in the present embodiment, the formulation of the stock selection strategy is mainly divided into two parts, the screening condition and the sorting condition.
The screening conditions are just like a sieve, some filtering conditions are specified in a stock set, the qualified filtering conditions are reserved as results, and the unqualified filtering conditions are eliminated. In addition to the filtering condition, the recommendation algorithm needs to use a sorting condition for determining which one is preferentially selected among the stocks that simultaneously satisfy the filtering condition.
For example, a-Z26 stocks are given a certain screening condition such as "stock price > -10 m", resulting in result sets a, B, C, D, a-D closing prices of 40, 30, 10, 20, respectively. And then setting the sorting condition as sorting according to the stock price from high to low, wherein the priority order of the recommended individual stocks is A > B > D > C, namely the individual stocks are sequentially pushed according to the order of A-B-D-C in actual operation. (the screening conditions and the sorting conditions in this example are only simple examples, and actually would be a combination of conditions.)
In addition, the stock selection strategy needs to consider several risk factors at the same time: individual stock liquidity, non-systematic risk (debarkation, pledge, business risk), i.e., the predetermined risk factors include, but are not limited to, individual stock liquidity, non-systematic risk (debarkation, pledge, business risk), etc.
And performing historical data back test on the stock selection strategy on the basis.
In this embodiment, the backtesting model performs historical backtesting on the second stock data according to a configurable first backtesting strategy to obtain a first historical backtesting result; and evaluating the first historical retest result according to at least one evaluation index, and correspondingly adjusting and updating the screening condition or/and the sequencing condition until the retest model is switched from the retest state to the working state when the first historical retest result output by the retest model meets a first expected condition.
In this embodiment, as shown in fig. 4, the retest model preferably employs a periodic rotation model. In the periodic rotation model, simulation conditions such as a bin adjustment period, a bin adjustment time point, a maximum bin holding quantity, a maximum bin position of each strand and the like need to be determined.
Wherein, as shown in fig. 5, the first test strategy comprises a plurality of combinations of test-back time, revenue benchmark, transaction cost and exclusion time period.
In this embodiment, the evaluation index includes a plurality of combinations of a total income evaluation index, a preset key evaluation index, and a positive correlation evaluation index; the preset key evaluation indexes comprise various combinations of sharp rate, maximum withdrawal rate, executable rate and income fluctuation rate.
In this embodiment, one way of evaluating the first historical return result according to at least one evaluation index includes:
judging whether the total income of the first historical return test result is greater than a total income evaluation index: if not, judging that the first historical return test result does not accord with the evaluation index;
if yes, continuously judging whether the key evaluation index of the first historical return test result meets a preset key evaluation index:
if not, judging that the first historical return test result does not accord with the evaluation index;
if yes, continuously judging whether the ranking condition and the income of the first historical return test result meet positive correlation evaluation indexes:
if not, judging that the first historical return test result does not accord with the evaluation index.
And if so, judging that the retest result meets the evaluation index.
In this embodiment, the first expected condition includes: multiple combinations of predetermined time wins (e.g., next day, 3 day, 5 day wins), yield (how much the rise and fall), maximum fall, and maximum gauging back depth (typically gauging back depth not exceeding 7%).
By backtesting the historical data of stock trading over a longer historical period, it is verified that a daily-based stock-selection strategy can theoretically yield positive benefits and properly combat non-systematic risks.
FIG. 6 shows historical data of a stock strategy based on the sun for a short line on a line. As shown in fig. 6, in the 5-year survey interval, the average profitability of the key indexes is 0.88%, the profit and profit are 52.19%, the profit and loss ratio is 5.88% to-4.57%, namely 1.287:1, and the executable proportion of the bin-adjusting instructions reaches 98.1% (almost no unexecutable transactions), and all indexes show that positive profit can be obtained on the stock-selecting strategy.
In this embodiment, after the review model enters the working state, the recommendation model selects, according to a configurable real-time push policy, a third stock of ticket data that meets the real-time recommendation condition from the second stock of ticket data to push and output.
Specifically, in this embodiment, as shown in fig. 7, the real-time pushing policy in the recommendation model includes:
calculating one or more push times for pushing the stock;
and judging whether stocks meeting the pushing conditions in the real-time disk exist in second stock data of the day before the day, if so, pushing the stocks to the user terminal, and if not, selecting corresponding stocks according to the ordering of the stocks in the second stock data to push the stocks to the user terminal when the pushing time is up.
When the conditions required by pushing in the disk are met (such as the real-time volume ratio of the current day is greater than 1.5), directly pushing a stock; when the conditions required by pushing in the disk are not met all the time, when the forced pushing time point calculated in advance arrives, the stocks with the highest scores are pushed out according to the sorting conditions (scores).
In this embodiment, after opening the stock on the second day, the corresponding stocks are sequentially selected from the stock pool in time-sharing mode to be pushed, so that the risk problem is mainly solved, firstly, the stocks can be recommended in time-sharing mode, the condition that the stock trend is influenced due to the fact that users buy the stocks too intensively and consistently is avoided, secondly, the combination condition of the stock-crossing data is difficult to manually judge which stock is recommended in a practical experience due to the fact that the recommendation rule in the stock pool is a series of time-sharing mode, and the A stock is a transaction mechanism of T +1, so that the risk that internal related personnel buy the stocks in advance to establish the rat chamber can be avoided by sequentially selecting the corresponding stocks from the stock pool in time-sharing mode after opening the stock on the second day.
Specifically, in the present embodiment, whether to execute the real-time push policy is controlled by a push probability function satisfying the push condition in the real-time disk. The push probability function is a monotonous function of the real-time comprehensive expansion amplitude, the distance market-receiving time and the push probability of each strand.
Suppose that the set latest recommended time is T0If the current time is T, then T0-T is the latest recommended time Δ T, and when Δ T is 0, selecting the corresponding stock from the second stock data and pushing the stock to the user terminal; and when the real-time comprehensive amplitude of each strand is smaller than the amplitude limit value, stopping executing the real-time pushing strategy.
The probability P that a stock is pushed by satisfying the push condition in real-time disk can be expressed as a function: p ═ F (T, R), there are two arguments:
1) the time is set to be T from the disk opening time, and T > is 0.
2) And (4) large-disc representation, setting real-time comprehensive expansion of two market stocks as R.
The disc opening time is set as a, and the disc closing time is set as b. The threshold for stopping pushing of R is S, and the graph shown in fig. 8 can be obtained.
The closer to market, the larger T and the larger P. As R becomes larger (the better the market), the criteria of the push requirement are more easily satisfied, and P rises. Conversely, when R becomes smaller, the market becomes worse and P decreases. When R < ═ S, the push stops completely. Therefore, in the ab interval, P and T, R have a monotonically increasing relationship, i.e. P is a monotonicity function.
Monotonicity interpretation of the function: monotonicity (monotonicity) of a function is also called as the increase and decrease of the function, and can qualitatively describe the relationship between the change of a function value and the change of an independent variable in a specified interval. When the argument of the function f (x) increases (or decreases) within its defined interval, the function value also increases (or decreases), and the function is said to be monotonous (monotonically increasing or monotonically decreasing) over that interval.
In particular, in this embodiment, the backtesting model performs a backtesting on the third stock of ticket data according to a configurable second backtesting policy to obtain a second historical backtesting result; the retest model judges whether the second historical retest result meets a second expected condition, and if so, the retest model does not act; and if not, the retest model is switched to the retest state from the working state when the state switching condition is met.
In this embodiment, the second retest strategy is as follows:
calculating the minimum price of each stock in the stock pool on the current recommended day, comparing the minimum price with the closing price of the previous day corresponding to the recommended day, and outputting a comparison result, namely a second historical return result;
and confirming whether the income obtained by buying the stock in the disk meets a second expected condition according to the second historical return test result, wherein the second expected condition is but not limited to: the probability of whether the profit gained by buying the stock in the disk is larger than the price of the stock in the second stock data selected according to the stock-selecting strategy; if so, a second expected condition is met; if not, the second expected condition is not satisfied.
As shown in fig. 9, the push time is a plurality of times, but not limited to, morning, noon, and evening hours. It should be noted that since the early disc distance is short per day, especially the last few minutes of disc opening such as volume ratio, and the fluctuation of the drive data such as buying, selling and the like is very large, if the drive data is used in the disc recommendation rule, the recommendation is generally prone to be performed after the disc is opened for a certain time such as 15 minutes; the luncheon disk has enough data volume and does not need to be carried out after 15 minutes. The recommended opportunity is therefore a difference between the morning and lunch recommendation rules.
In this embodiment, the main mechanism of the push condition in the real-time disk is shown in fig. 7. After opening the disc, one or several time points are fixed. And (4) pushing stocks which meet the real-time pushing condition before the time point. If not, pushing according to the sorting condition when reaching the time point.
In this embodiment, the real-time in-disk push condition is mainly based on the following factors: index rise and fall amplitude, individual strand rise and fall amplitude, disc opening information and the like. The purpose of judging the index fluctuation amplitude is to partially avoid systematic risks and stop stock recommendation when the index has a small-probability fluctuation event.
Generally, historical data retrieval of a stock selection rule in a tray based on individual stock fluctuation amplitude and tray opening information conditions needs to be based on a large amount of historical time-sharing data, and the method is very difficult to realize. However, the angle of the problem can be changed, and in the case that the data of the stock-selecting strategy return measurement based on the daily line is ideal, only the lowest price of the recommended day needs to be calculated and compared with the previous closing price of the recommended day, so as to confirm how much probability of buying in the disk can obtain the price which is better than the strategy based on the daily line.
In this example, the half-year data of 2019/6/3-2019/12/3 after the fact that a strategy stock pool is online is used for verification, and it is found that in 405 recommended records, 375 records have the lowest price of the following 1 trading day lower than the closing price of the day before recommendation, and account for 92.6% of the total proportion of all samples. It can be considered that push-in-disk does not significantly reduce profitability.
The stock transaction is divided into three links of buying, holding and selling, and the processes of the method only surround the link of buying. The remaining two important links, holding and selling, are manually judged by most individual investors. In order to provide a more objective and consistent operation methodology for users, it is also of positive significance to make explicit the ending of filling and losing while recommending stocks.
In this embodiment, the recommendation model pushes the determined ending surplus price and ending loss price while recommending the stocks to the user terminal; and the return test model carries out return test on the determined stop loss price and the stop loss price, and determines whether the currently pushed stop loss price and stop loss price are suitable.
Specifically, in this embodiment, one implementation manner of determining the stop filling price and the stop loss price includes:
respectively calculating the total number of profit trades and the total number of loss trades of the stock trading history data of the stock according to the stock selection strategy; respectively calculating the average income of the stock profit transaction and the average income of the loss transaction according to the total number of the profit transaction and the total number of the loss transaction; determining a loss value near 1 standard deviation according to the distribution of the maximum loss of all stocks, determining the loss value as a reserve price, and reversely pushing a minimum reserve value required for realizing profit according to a profit-loss ratio, and determining the minimum reserve price.
For example, there are 1000 transactions in a given time period, with 400 profits, 600 losses, and a profit-loss ratio of 40/60-1: 1.5. The average profit of 400 profit transactions is 10%, the average profit of 600 loss transactions is-5%, theoretically the total profit is 400 × 10% +600 × 5% — 1000%, and the average profit of a single transaction is 1000%/1000 ═ 1%.
According to the calculated profit-loss ratio of 1:1.5, if profit is to be realized, the filling/loss should be at least greater than 1.5, i.e. the filling should be at least greater than 1.5 x% at the loss-stop position.
Table 1 is the share of a stock-picking strategy where the deficit reaches the specified value within 5 days.
As shown in fig. 10, 31.2% of the stocks in the sample reached-4%, i.e., 68.8% of the stocks did not reach-4% of the stocks in the 5-day period, and was just around 1 standard deviation (if the statistics were normal distribution). It is reasonable to set the stop loss at-4%. Meanwhile, according to the profit-loss ratio, the thrust must exceed 6%, and is assumed to be 7%.
In this embodiment, for example, after the ending loss is determined, the final profit for all stocks is controlled to be-4% -7%. And finally, calculating the average profit according to the recalculated final profit to finally determine whether the profit can be realized by the strategy after the filling and the loss stopping are added.
The natural rate of return and the rate of return distribution map 11 and fig. 12 for setting the stop-filling and stop-loss are shown with 5 days as a trading period, and the rate of return is converged to some extent.
Therefore, the stock selection strategy of the stock pool in the embodiment is supported by historical backtesting, the probability of profit is high, corresponding stocks are selected from the stock pool and pushed to the user terminal in time intervals in sequence after the stock pool is opened on the second day, and the stocks can be recommended in time intervals, so that the situation that the stock buying is too concentrated and consistent by the user and the stock buying trend is influenced is avoided.
The embodiment also provides an intelligent stock recommendation method based on the remeasurement data, which comprises the following steps: a step of back test and a step of real-time push.
1) The back measurement step comprises:
screening the current stock data according to a configurable screening strategy, and outputting first stock data meeting screening conditions; sorting the first stock data meeting the screening condition according to a configurable sorting strategy, and outputting second stock data which meet the sorting condition and are sequentially arranged; performing historical backtesting on the second stock data according to a configurable first backtesting strategy to obtain a first historical backtesting result; and evaluating the first historical retest result according to at least one evaluation index, and correspondingly adjusting and updating the screening condition or/and the sequencing condition until the retest model is switched from the retest state to the working state when the first historical retest result output by the retest model meets a first expected condition.
2) The real-time pushing step comprises the following steps:
after the retest model enters a working state, selecting third stock of ticket data meeting real-time recommendation conditions from the second stock of ticket data according to a configurable real-time pushing strategy to carry out pushing output; carrying out retesting on the third stock of ticket data according to a configurable second retesting strategy to obtain a second historical retesting result; judging whether the second historical retest result meets a second expected condition, if so, not acting; and if not, the retest model is switched to the retest state from the working state when the state switching condition is met.
Technical features of specific implementation of the stock intelligent recommendation method based on the measured data in this embodiment are substantially the same as those of the stock intelligent recommendation system based on the measured data in the foregoing embodiment, and common technical contents among the embodiments are not repeated.
As shown in fig. 13, a schematic structural diagram of the electronic terminal 102 in the embodiment of the present application is shown.
The electronic terminal 102 includes:
at least one display 1001 for displaying. In this embodiment, the display may be an OLED, LED, LCD display, or the like.
At least one memory 1002 for storing computer programs;
at least one processor 1003, coupled to the display 1001 and the memory 1002, is configured to run the computer program to implement the steps of the above-mentioned stock intelligent recommendation method based on the measured data.
The memory 1102 is connected with the processor 1101 through a system bus and is used for completing mutual communication, the memory 1102 is used for storing computer programs, and the processor 1101 is used for operating the computer programs, so that the electronic terminal executes the stock intelligent recommendation method based on the return survey data. The above-mentioned stock intelligent recommendation method based on the survey data has been described in detail, and is not repeated herein.
It should be noted that the above-mentioned system bus may be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The system bus may be divided into an address bus, a data bus, a control bus, and the like. For ease of illustration, only one thick line is shown, but this does not mean that there is only one bus or one type of bus. The communication interface is used for realizing communication between the database access device and other equipment (such as a client, a read-write library and a read-only library). The Memory may include a Random Access Memory (RAM), and may further include a non-volatile Memory (non-volatile Memory), such as at least one disk Memory.
The Processor 1101 may be a general-purpose Processor, and includes a Central Processing Unit (CPU), a Network Processor (NP), and the like; the device can also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, a discrete Gate or transistor logic device, or a discrete hardware component.
Those of ordinary skill in the art will understand that: all or part of the steps for implementing the above method embodiments may be performed by hardware associated with a computer program. The aforementioned computer program may be stored in a computer readable storage medium. When executed, the program performs steps comprising the method embodiments described above; and the aforementioned storage medium includes: various media that can store program codes, such as ROM, RAM, magnetic or optical disks.
In this embodiment, the user terminal is a terminal used by a user, and the electronic terminal 102 is a terminal used by a company providing stock service data. The user terminal displays a client service graphical interface providing stock service data, and the client service graphical interface is accessed by a webpage browsed by a browser loaded by the user terminal; wherein the user terminal comprises a PC; or the customer service graphical interface is accessed by an interface provided by integrated service platform software loaded by the user terminal; the user terminal comprises a mobile terminal and a PC; the mobile terminal comprises a smart phone or a tablet computer; the integrated service platform software comprises WeChat and/or Payment treasures.
It should be noted that the displayed graphical user interface may also be different for different types of user terminals.
Specifically, in some embodiments, for the user terminal to be a PC terminal, it may browse a web page through a loaded browser (including but not limited to IE, google, 360, QQ, dog search, hundredth, standing swim, UC, fox fire, cheetah, 2345, punish, etc.) and interface with the web page by accessing a predetermined URL to access a specific web page provided by the service terminal 102, in which a graphical interface of the customer service is displayed.
In yet other embodiments, for the user terminal to be a mobile terminal (e.g., a smart phone, a tablet computer), the user terminal may access the graphical interface of the customer service through a web page or an applet in the integrated platform software, such as WeChat, Paibao, etc.
The registered user can add the wechat applet in a mode of scanning the two-dimensional code or searching the wechat applet, and then operate (for example, click) the wechat applet so as to enter the customer service graphical interface.
In one embodiment, the user terminal may be, for example, a fixed terminal, such as a desktop computer; or a mobile terminal, such as a notebook computer, a smart phone, or a tablet computer.
The user terminal is provided with a client side APP, a client can log in the client side APP through pre-registered account information, the client side APP can authenticate by self, and stock service data related to the account information is provided after the authentication is passed.
Optionally, the user terminal and the electronic terminal 102 may implement data communication via a C/S architecture, that is, the user terminal installs client software, and the client software may request the electronic terminal 102 to download the stock service data.
Preferably, the data communication between the user terminal and the electronic terminal 102 can be implemented by a B/S architecture, that is, the user terminal is provided with a Browser (Browser) for displaying the graphical interface of the client service. Based on the B/S framework, the requirements for the hardware and software of the user terminal of the client can be greatly reduced, the user terminal does not need to install client software and only needs to be provided with a web browser, and therefore the user experience of the client can be greatly improved.
In summary, in the present embodiment, the stock intelligent recommendation system and method based on the backtesting data and the electronic terminal perform historical backtesting on the stock selection policy, so that a high profit probability is achieved, the configurable real-time push policy and the timesharing recommendation are used to avoid that the stock trends are affected due to too concentrated and consistent buying of the user, and the pushed stocks are continuously backtested, so that more objective and highly accurate stocks are effectively recommended to the user, and a suitable profit-loss ratio is set, so that the maximization of the relative profitability is achieved.
. Therefore, the stock intelligent recommendation system and method based on the backtesting data and the electronic terminal have higher innovativeness compared with the prior art.
The above embodiments are merely illustrative of the principles and utilities of the present application and are not intended to limit the application. Any person skilled in the art can modify or change the above-described embodiments without departing from the spirit and scope of the present application. Accordingly, it is intended that all equivalent modifications or changes which can be made by those skilled in the art without departing from the spirit and technical concepts disclosed in the present application shall be covered by the claims of the present application.
Claims (10)
1. The utility model provides a stock intelligence recommendation system based on survey data which characterized in that: the method comprises the following steps:
the screening model is used for screening the current stock data according to a configurable screening strategy and outputting first stock data meeting screening conditions;
the sorting model is used for sorting the first stock data meeting the screening conditions according to a configurable sorting strategy and outputting second stock data which meet the sorting conditions and are arranged in sequence;
the backtesting model is used for performing historical backtesting on the second stock data according to a configurable first backtesting strategy to obtain a first historical backtesting result;
evaluating the first historical retest result according to at least one evaluation index, and correspondingly adjusting and updating the screening condition or/and the sequencing condition until the retest model is switched from the retest state to the working state when the first historical retest result output by the retest model meets a first expected condition;
the recommendation model selects third stock data meeting the real-time recommendation condition from the second stock data according to a configurable real-time push strategy to carry out push output after the retest model enters a working state;
the backtesting model performs backtesting on the third stock of ticket data according to a configurable second backtesting strategy to obtain a second historical backtesting result;
the retest model judges whether the second historical retest result meets a second expected condition, and if so, the retest model does not act; and if not, the retest model is switched to the retest state from the working state when the state switching condition is met.
2. The intelligent stock recommendation system based on the return survey data as claimed in claim 1, wherein: the first reward strategy comprises a plurality of combinations of reward time, a revenue benchmark, a transaction cost and an exclusion time period; the evaluation indexes comprise various combinations of total income evaluation indexes, preset key evaluation indexes and positive correlation evaluation indexes; the preset key evaluation indexes comprise various combinations of sharp rate, maximum withdrawal rate, executable rate and income fluctuation rate.
3. The intelligent stock recommendation system based on the measured data according to claim 1 or 2, characterized in that: the first expected condition includes: a plurality of combinations of a win rate, a profit rate, a maximum fall and a maximum return depth for a preset time.
4. The intelligent stock recommendation system based on the return survey data as claimed in claim 1, wherein: the screening strategy comprises a configuration strategy for one or more combinations of the market value, the time interval and the market value; the ranking strategy comprises a configuration strategy of deducting one or more combinations of net asset profitability, order, range and weight.
5. The intelligent stock recommendation system based on the return survey data as claimed in claim 1, wherein: the real-time push strategy in the recommendation model comprises the following steps:
calculating one or more push times for pushing the stock;
and judging whether stocks meeting the pushing conditions in the real-time disk exist in second stock data of the day before the day, if so, pushing the stocks to the user terminal, and if not, selecting corresponding stocks according to the ordering of the stocks in the second stock data to push the stocks to the user terminal when the pushing time is up.
6. The intelligent stock recommendation system based on the measured data according to claim 1 or 5, characterized in that: controlling whether to execute a real-time push strategy or not through a push probability function meeting push conditions in a real-time disk; the push probability function is a monotonous function of the real-time comprehensive expansion amplitude, the distance market receiving time and the push probability of each strand;
suppose that the set latest recommended time is T0If the current time is T, then T0-T is the latest recommended time Δ T, and when Δ T is 0, selecting the corresponding stock from the second stock data and pushing the stock to the user terminal; and when the real-time comprehensive amplitude of each strand is smaller than the amplitude limit value, stopping executing the real-time pushing strategy.
7. The intelligent stock recommendation system based on the return survey data as claimed in claim 1, wherein: the recommendation model also pushes the determined filling price and loss price while recommending the stocks to the user terminal;
and the return test model carries out return test on the determined stop loss price and the stop loss price, and determines whether the currently pushed stop loss price and stop loss price are suitable.
8. The intelligent stock recommendation system based on the return survey data as claimed in claim 7, wherein: one implementation of determining the reserve price and the reserve price includes:
respectively calculating the total number of profit trades and the total number of loss trades of the stock trading history data of the stock according to the stock selection strategy;
respectively calculating the average income of the stock profit transaction and the average income of the loss transaction according to the total number of the profit transaction and the total number of the loss transaction;
determining a loss value near 1 standard deviation according to the distribution of the maximum loss of all stocks, determining the loss value as a reserve price, and reversely pushing a minimum reserve value required for realizing profit according to a profit-loss ratio, and determining the minimum reserve price.
9. An intelligent stock recommendation method based on backtesting data is characterized in that: the method comprises the following steps:
and (3) a retesting step:
screening the current stock data according to a configurable screening strategy, and outputting first stock data meeting screening conditions;
sorting the first stock data meeting the screening condition according to a configurable sorting strategy, and outputting second stock data which meet the sorting condition and are sequentially arranged;
performing historical backtesting on the second stock data according to a configurable first backtesting strategy to obtain a first historical backtesting result;
evaluating the first historical retest result according to at least one evaluation index, and correspondingly adjusting and updating the screening condition or/and the sequencing condition until the retest model is switched from the retest state to the working state when the first historical retest result output by the retest model meets a first expected condition;
a real-time pushing step:
after the retest model enters a working state, selecting third stock of ticket data meeting real-time recommendation conditions from the second stock of ticket data according to a configurable real-time pushing strategy to carry out pushing output;
carrying out retesting on the third stock of ticket data according to a configurable second retesting strategy to obtain a second historical retesting result; judging whether the second historical retest result meets a second expected condition, if so, not acting; and if not, the retest model is switched to the retest state from the working state when the state switching condition is met.
10. An electronic terminal, characterized by: the method comprises the following steps:
at least one display for displaying;
at least one memory for storing a computer program;
at least one processor, connected to the display and the memory, for executing the computer program to implement the method for intelligently recommending stocks based on the measured data as claimed in claim 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010642346.7A CN111815447A (en) | 2020-07-06 | 2020-07-06 | Stock intelligent recommendation system and method based on backtesting data and electronic terminal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010642346.7A CN111815447A (en) | 2020-07-06 | 2020-07-06 | Stock intelligent recommendation system and method based on backtesting data and electronic terminal |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111815447A true CN111815447A (en) | 2020-10-23 |
Family
ID=72843030
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010642346.7A Pending CN111815447A (en) | 2020-07-06 | 2020-07-06 | Stock intelligent recommendation system and method based on backtesting data and electronic terminal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111815447A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112734577A (en) * | 2021-03-01 | 2021-04-30 | 方玉龙 | Multicycle decision-making AI stock-frying robot platform |
| CN112948700A (en) * | 2021-04-14 | 2021-06-11 | 刘蒙 | Fund recommendation method |
| CN112966189A (en) * | 2021-04-14 | 2021-06-15 | 刘蒙 | Fund product recommendation system |
| CN116739789A (en) * | 2023-08-16 | 2023-09-12 | 中信证券股份有限公司 | Virtual article return information sending method and device, electronic equipment and medium |
| CN117455250A (en) * | 2023-10-10 | 2024-01-26 | 上海卡方信息科技有限公司 | Service execution method, device, equipment and readable storage medium |
| TWI842415B (en) * | 2023-03-01 | 2024-05-11 | 康和綜合證券股份有限公司 | Device and method for filtering transaction data, and computer program product thereof |
-
2020
- 2020-07-06 CN CN202010642346.7A patent/CN111815447A/en active Pending
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112734577A (en) * | 2021-03-01 | 2021-04-30 | 方玉龙 | Multicycle decision-making AI stock-frying robot platform |
| CN112948700A (en) * | 2021-04-14 | 2021-06-11 | 刘蒙 | Fund recommendation method |
| CN112966189A (en) * | 2021-04-14 | 2021-06-15 | 刘蒙 | Fund product recommendation system |
| CN112966189B (en) * | 2021-04-14 | 2024-01-26 | 北京基智科技有限公司 | Fund product recommendation system |
| TWI842415B (en) * | 2023-03-01 | 2024-05-11 | 康和綜合證券股份有限公司 | Device and method for filtering transaction data, and computer program product thereof |
| CN116739789A (en) * | 2023-08-16 | 2023-09-12 | 中信证券股份有限公司 | Virtual article return information sending method and device, electronic equipment and medium |
| CN116739789B (en) * | 2023-08-16 | 2023-12-19 | 中信证券股份有限公司 | Virtual article return information sending method and device, electronic equipment and medium |
| CN117455250A (en) * | 2023-10-10 | 2024-01-26 | 上海卡方信息科技有限公司 | Service execution method, device, equipment and readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111815447A (en) | Stock intelligent recommendation system and method based on backtesting data and electronic terminal | |
| Clegg et al. | Pairs trading with partial cointegration | |
| US10453142B2 (en) | System and method for modeling and quantifying regulatory capital, key risk indicators, probability of default, exposure at default, loss given default, liquidity ratios, and value at risk, within the areas of asset liability management, credit risk, market risk, operational risk, and liquidity risk for banks | |
| JP2001525577A (en) | Financial advisory system | |
| US20140258175A1 (en) | Generating Personalized Investment Recommendations | |
| JP2005530232A (en) | System and method for transaction cost assessment and optimization | |
| AU2005250958A1 (en) | A system and method for analysing risk associated with an investment portfolio | |
| Shah et al. | Principal component analysis based construction and evaluation of cryptocurrency index | |
| Cullen et al. | Changes to mutual fund risk: Intentional or mean reverting? | |
| Ssebunya et al. | Group membership and certification effects on incomes of coffee farmers in Uganda | |
| Wang et al. | Can investors’ informed trading predict cryptocurrency returns? Evidence from machine learning | |
| CN113159972B (en) | Combination determination method, device, electronic equipment and computer readable storage medium | |
| Chen et al. | Household tenure choice and housing price volatility under a binding home-purchase limit policy constraint | |
| CN110175883A (en) | Sorting method, sorting device, electronic equipment and nonvolatile storage medium | |
| Wei et al. | Profit seeking vs. survival seeking: An analytical study of supplier's behavior and buyer's subsidy strategy | |
| KR20120032606A (en) | Stock investment system enabling participattion of stock investment clients and method thereof | |
| Banholzer et al. | Exploiting investor sentiment for portfolio optimization | |
| Makkonen et al. | Multi‐criteria decision support in the liberalized energy market | |
| US20210056625A1 (en) | Artificial intelligence and machine learning in a clustering processes to develop a utility model for asset allocation and automated rebalancing of exchange traded funds | |
| US20190279301A1 (en) | Systems and Methods Using an Algorithmic Solution for Analyzing a Portfolio of Stocks, Commodities and/or Other Financial Assets Based on Individual User Data to Achieve Desired Risk Based Financial Goals | |
| CN112750059A (en) | Evaluation method of industry support policy, processing method of industry support result and related product | |
| Kwok | Macroeconometric models for portfolio management | |
| Crawford et al. | Automatic High‐Frequency Trading: An Application to Emerging Chilean Stock Market | |
| Ghuloum et al. | Joint dynamic pricing with acquisition and selling opportunities | |
| WO2001027844A9 (en) | Investment analysis and management system for generating financial advice |
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 |