TW201510905A - Strategic trading method - Google Patents

Abstract

A strategic trading method includes using a processor to provide an initial intensity formula. The initial intensity formula includes a plurality of primary determinants and a plurality of weighting factors corresponding to the primary determinants. The strategic trading method further includes using the processor to optimize the weighting factors so as to generate an updated intensity formula according to previous trading data.

Description

Strategic trading method

The present invention relates to a strategic trading method, and more particularly to a strategic trading method for providing investment advice based on the strength information of buying and selling.

In the previous strategy trading methods, such as the use of the investment assistance software available on the market, the following functions have been used: using the processor and the market open database to automatically analyze the market data and corresponding multiple statistical line types, and pre-set The user is notified when the preset condition is met. For example, when a user uses an interactive investment assistance software that is connected to the stock market open database, the user can set in advance when one or more preset conditions are met, for example, the legal person buys for three consecutive days. When a specific number of shares are exceeded, the processor is instructed to send a signal to the user's mobile device to notify the user that the preset condition has been met. In another example of the prior art, the user can also set a statistical line type obtained according to statistical data related to a particular stock or market trend. When a typical line type is met, the processor sends a reminder signal to the use. The mobile device. For example, when the specific line of the Candlestick chart and the classic K-line analysis textbook stored in the investment assistance software database are the same as the recommended "Red Three Soldiers" line type, the processor can recognize The K-line type "Red Three Soldiers" has already occurred. If the user has previously set the "Red Three Soldiers" line type to appear as a preset condition, the processor will issue a notification. For another example, in the prior art, the user can monitor the stock price of a particular stock, and when one or a plurality of predetermined preset conditions specified in advance, such as "five minutes all the way up" or "downstop lock" occurs, The processor can recognize the occurrence of the conditions and send a signal to the user's mobile device to notify the user. The above examples of prior art use the stock market as an example, but the prior art can also assist The user grasps the price changes and statistical linear changes of other investment fields such as foreign exchange markets or futures trading, and when the changes meet the preset conditions, the processor sends a signal to notify the user. Please refer to FIG. 1 , which is a setting interface of preset conditions in the prior art. As can be seen from Figure 1, the user can select "Notify Me" after reviewing the various conditions provided by the system, and then receive a notification sent by the processor when the condition is met.

In the face of the complicated price changes in the market, various linear changes and other factors that need to be considered in many investment decisions, the prior art can only set the preset conditions one by one, and then judge whether the preset conditions occur by the processor. Therefore, the prior art can only inform the user when various preset conditions occur. For users who attempt to make more detailed and comprehensive considerations of market information, the prior art cannot help the user to consider the comprehensive effects caused by various market changes, which is too simplified and Fragmented. The prior art is also unable to present the main decisions of the investment decision, such as price kinetic energy, fundamentals, technical, chip, situation and/or pressure support, the impact and importance of the investment results. Therefore, the prior art as a method of assisting the user in monitoring the market has the disadvantage of being too fragmentary and too simplistic.

An embodiment of the present invention discloses a policy transaction method, including using a processor to provide a starting strength equation for adjusting a portfolio, the starting strength equation including a plurality of main decision facets and a plurality of corresponding weight factors, and using the The processor optimizes the plurality of weighting factors based on previous transaction data to generate an updated intensity equation to adjust the portfolio.

201‧‧‧ processor

202‧‧‧Market open database

203‧‧‧Database linked to the processor

204‧‧‧Handheld electronic devices and/or one-person computers

401, 402, 403‧‧ steps

Fig. 1 is a setting interface of a preset condition in the prior art.

Fig. 2 is a schematic view showing the arrangement of an embodiment of the present invention.

The third picture shows the typical line type in the K-line analysis textbook "The Law of the Wine Field".

Figure 4 is a flow chart of the method of the present invention.

The following is a detailed description of the present invention in accordance with the present invention, and the embodiments are not intended to limit the scope of the invention.

Please refer to Figure 2. Fig. 2 is a schematic view showing the arrangement of an embodiment of the present invention. As shown in FIG. 2, the embodiment includes a processor 201, a market disclosure database 202, a database 203 coupled to the processor, and a handheld electronic device and/or a personal computer 204. The processor 201 may retrieve the data of the face factor corresponding to the plurality of main decision facets from the market disclosure database 202, calculate the buy signal strength I, and further provide the adjusted investment portfolio according to the purchase signal strength I. It is recommended to the user's handheld electronic device and/or personal computer 204. The database 203 connected to the processor pre-stores various line types of a specific mode and is used as the processor 201 for comparison. The main decision to purchase the face can be the price kinetic energy surface, the fundamental surface, the technical surface, the chip surface, the situation surface, the pressure support surface and the like. Each main decision purchase contains at least one facet factor, for example, the fundamentals include face-to-face factors such as monthly revenue growth rate, earnings per share growth rate and gross profit margin growth rate. The above suggestions for adjusting the investment portfolio according to the buy signal strength I may, for example, adjust the buy and sell strategy, the stop loss stop strategy and the fund utilization strategy according to the buy signal strength I.

The embodiment of FIG. 2 discloses a policy transaction method including providing a starting strength equation using a processor 201, the starting strength equation including a plurality of main decision facets and a plurality of corresponding weight factors, and using the processor 201 according to the previous The transaction data optimizes the plurality of weighting factors to produce an updated intensity equation. The initial intensity equation is as follows: ;among them

Available after expansion Wherein I represents the buy signal strength; X ₁ to X _n each represent the 1st to nth main decision facets, where n is a positive integer and n>1; W _a represents a corresponding to X _a , ie the a-th main decision The weighting factor of the facet, where a is a positive integer and n≧a≧1; and X _ai ie X _a1, X _a2, X _a3 ... represent the facet factor contained in the a-th main decision facet, since each The number of facet factors included in the main decision facet is not fixed, so There is no upper limit in the formula.

In the above formula (A), the buy signal strength I is a total score obtained by weighting the calculation for an investment item, such as a specific stock. The greater the value of the buy signal strength I, the higher the probability of successful trading, so the more it is recommended to buy the investment project, in addition, the purchase of funds can also be amplified. At the beginning of the calculation, the buy signal strength I has a preset starting value.

In the above formula, the weighting factors W ₁ to W _n corresponding to the first to nth main decision surface planes also each have a preset starting value at the start of the calculation. In the present invention, the a-th main decision facet X _{a is} weighted by the weighting factor W _a to adjust and influence the influence and importance of the main decision facet X _a on the buy signal strength I.

According to an embodiment of the present invention, when the main decision surface is the stock kinetic energy surface, the fundamental surface, the technical surface, the chip surface, the surface surface, and the pressure support surface, the description is as follows: The stock price can be composed of a plurality of facet factors, which can include the market value of the day, the market value of the day, the rise or fall of the market value, whether the stock price hits the current high and the high level of innovation, whether the stock price hits the current low and the low level of innovation. The trading volume of the day was several times of yesterday's trading volume, the trading volume of the day was several times of the average daily trading volume, the largest single trading volume of the day was several times of the largest single trading volume yesterday, and the estimated trading volume of the day was Several times the transaction volume of w, where u, v, and w are positive integers that can be set by the user, and each facet factor can be obtained by the market public database 202 and converted into a score.

For example, if the main decision facet "share price active face" is set to X ₁ and one of the multiple facet factors that constitute the active face of the stock price is the face factor "the daily trading volume is the average daily trading volume of v. "Multiply" is set to Y ₁₁ (v), and the score corresponding to Y ₁₁ (v) is set to X ₁₁ , and the value of v can be set by the user, and the processor 201 can obtain the daily transaction volume and the near market from the market disclosure database 202. v Average daily trading volume, calculate the daily trading volume as several times the average daily trading volume, and further convert it into a rating X ₁₁ . For example, when the user sets v=5, Y ₁₁ (5)=the current day volume is close to the average daily volume of 5 days, and then can be based on the built-in comparison stored in the processor 201 or the database 203 connected to the processor. Table, convert this result to a rating. The comparison table can be as follows, for example:

When Y ₁₁ (5)=the daily trading volume is close to the average daily trading volume of 5.03, the table can be found that the facet factor “the daily trading volume is several times the average daily trading volume”, the score X ₁₁ is 0 points. The calculation process is organized as follows: To consider the facet factor "the daily trading volume is several times the average daily trading volume", the user sets v=5; →Y ₁₁ (5)=the daily trading volume is nearly 5 days average transaction Quantity = 1.03; → look up the table; → get X ₁₁ =0; → return X ₁₁ =0 to the intensity equation (A).

This rating is used to indicate the extent and importance of the event represented by the facet factor.

For example, when Y ₁₁ (5) = the daily trading volume is close to the average daily trading volume of 5 days = 0.3, the table can be found that the facet factor "the daily trading volume is several times the average daily trading volume" X ₁₁ is - 1 point, the calculation process is organized as follows: To consider the facet factor "the daily trading volume is several times the average daily trading volume", the user sets v=5; →Y ₁₁ (5)=the daily trading volume is close to 5 Daily average volume = 1.03; → look up the table; → get X ₁₁ = -1; → return X ₁₁ = -1 to the intensity equation (A).

This rating is used to indicate the extent and importance of the event represented by the facet factor. In the same way, the remaining facet factors can also be obtained by the market public database 202, and then the corresponding scores are obtained through the calculation process and the look-up table method, respectively, to replace the intensity equation (A).

The fundamentals may consist of a plurality of facet factors, which may include monthly revenue growth, earnings per share growth, gross margin quarterly growth, and equity return quarterly growth, and each facet factor may be obtained from the market disclosure database 202 and Convert to a rating.

For example, if the main decision facet "Fundamental Plane" is set to X ₂ and one of the plurality of facet factors that constitute the fundamental face factor is "Year of earnings per share" is set to Y ₂₁ and The score corresponding to Y ₂₁ is set to X ₂₁ , and the processor 201 can obtain the earnings per share of the investment project and the earnings per share of the previous quarter from the market disclosure database 202, and calculate the corresponding score according to, for example, the following table:

When Y ₂₁ = earnings per share growth = earnings per share for the quarter/previous quarter earnings per share -1.33, the table can be found that the facet factor "per share earnings growth" score of ₂₁ is 0.5 points. The calculation process is organized as follows: To consider the facet factor "growth earnings per share"; → Y ₂₁ = earnings per share for the quarter/previous quarter earnings per share -1 = 1.33-1 = 0.33; → check the table; → get X ₂₁ =0.5;→Return X ₂₁ =0.5 back to the intensity equation (A).

This rating is used to indicate the extent and importance of the event represented by the facet factor.

For example, when Y ₂₁ = earnings per share growth = earnings per share for the quarter/previous quarter earnings per share -1 = 2.05-1 = 1.05, check the table to see the face factor "growth earnings per share growth" rating Y ₂₁ is 1 point, the operation process is organized as follows: To consider the facet factor "growth earnings per share"; → Y ₂₁ = earnings per share this quarter, earnings per share -1 = 1.05; → check the table; X ₂₁ =1; → return X ₂₁ =1 to the intensity equation (A).

This rating is used to indicate the extent and importance of the event represented by the facet factor. Similarly, the remaining facet factors can also be retrieved from the market open database 202, and then the corresponding scores are obtained through the calculation process and the look-up table method, respectively, to replace the intensity equation (A).

The technical plane may be composed of a plurality of facet factors, which may include a line type judgment of a long-term average of the closing average, a line type judgment of whether the KD line is greater than 80%, a line type judgment of whether the KD line is less than 20%, and whether the tower line is reddish. The line type judgment and the line type judgment of whether the pagoda line is blacked out, and each facet factor can be obtained by the market public database 202 and converted into a score.

For example, if the main decision facet "technical face" is set to X ₃ , and one of the plurality of facet factors that constitute the technical face, the facet factor "whether the pagoda line turns red" is set to Y ₃₁ , and The score corresponding to Y ₃₁ is set to X ₃₁ , and the processor 201 can obtain the price fluctuation range and the daily closing price of an investment item from the market disclosure database 202, and obtain the corresponding pagoda line type according to the following table. score:

The pagoda line is generated by daily closing price and daily price fluctuations, which are used to assist users in observing price trends rather than estimating the highest/lowest point of the price. When the stock price reverses from the bottom up, the black bar line of the pagoda line enters a reddish state, indicating that the price of the investment project begins to rise. In general, the user can consider the right amount of purchase at this time. Check the table to see if the facet factor "whether the pagoda line turns red" corresponds to the score Y ₃₁ is 1 or 0. The calculation process is as follows: To consider the facet factor "whether the pagoda line turns red"; → processor 201 to the market The public database 202 obtains the daily closing price and the price fluctuation of an investment project, and draws the pagoda line after the statistics; → the line type of the pagoda that has been drawn, and the pre-stored in the database 203 connected to the processor. Whether the pagoda line is red or not, the line types are compared with each other to know whether the pagoda line is turned red by black; → check the table; → if the pagoda line turns red from black, it means that the face factor “whether the pagoda line turns red” is judged as "Yes", indicated by Y ₃₁ =1, and the corresponding score X ₃₁ =1; if the pagoda line does not turn red, it means that the facet factor "whether the pagoda line turns red" is judged as "No" to Y ₃₁ =0 indicates that the corresponding score X ₃₁ =0; → returns X ₃₁ back to the intensity equation (A).

This rating is used to indicate the extent and importance of the event represented by the facet factor. In the same way, the remaining facet factors can also be obtained by the market public database 202, and then the corresponding scores are obtained through the calculation process and the look-up table method, respectively, to replace the intensity equation (A).

The chip surface may be composed of a plurality of facet factors, which may include whether the legal person has a net purchase over the continuous r day, whether the legal person has a net sales over the s day, whether the legal person's shareholding ratio is greater than p%, and whether the legal person's shareholding ratio is less than q%, A score indicates that r and s are positive integers that can be set by the user, and p and q are positive numbers that can be set by the user, where 0 ≦ p < 100 and 0 < q ≦ 100, and each face factor Both can be obtained by the market disclosure database 202 and converted into a rating.

For example, if the main decision facet "chip face" is set to X ₄ and one of the plurality of facet factors that make up the chip face, the face factor "whether the legal person is continuous r day net buy super" is set to Y ₄₁ (r), and the score corresponding to Y ₄₁ (r) is set to X ₄₁ , and the value of r can be set by the user, the processor 201 can obtain the purchase and sale of the legal person for the continuous r day from the market disclosure database 202. According to the judgment of the facet factor "whether the legal person continues to buy the net for the continuous r day", and further convert it into a score X ₄₁ . For example, when the user sets r=3, Y ₄₁ (3) indicates “whether the legal person has a net purchase for 3 consecutive days”, and then can be based on the built-in comparison stored in the processor 201 or the database 203 connected to the processor. Table, convert this result to a rating. The comparison table can be as follows, for example:

When Y ₄₁ (3) = whether the legal person has bought the net for 3 consecutive days, if the face factor "whether the legal person has a net purchase of 3 consecutive days" has occurred, then Y ₄₁ (3) = 1 and the corresponding score X ₄₁ is 1 point. The calculation process is organized as follows: To consider the facet factor "Whether the legal person continuously buys the net on the r day", the user sets r=3; →Y ₄₁ (3)=Whether the legal person has bought the net for 3 consecutive days → If the legal person has For the third consecutive day, the net purchase is over, then Y ₄₁ (3) = 1; → look up the table; → get X ₄₁ =1; → return X ₄₁ =1 to the intensity equation (A).

This rating is used to indicate the extent and importance of the event represented by the facet factor. Similarly, the remaining facet factors can also be retrieved from the market open database 202, and then the corresponding scores are obtained through the calculation process and the look-up table method, respectively, to replace the intensity equation (A).

The situation can be composed of a plurality of facet factors, which can include the strong indicators of ethnic stocks, the global stocks strong indicators, the trend line type judgment, the K line line judgment and the wine field line type judgment. Among them, the strong indicators of ethnic stocks are the percentage increase of the group stocks compared with the increase of the broader market, and the global stocks strong indicators are the percentage of the global stocks compared with the increase of the broader market. In addition, the trend line type judgment, the K line line type judgment and the wine field line type judgment are the trend line type, the K line line type and the wine field line type which are obtained by the statistics obtained by the market open database 202, respectively, and the database linked to the processor. The line patterns of the specific patterns pre-stored in 203 are compared with each other, and it is judged whether the line type obtained after the statistics conforms to the line type of the pre-stored specific mode. Each facet factor can be taken by the market disclosure database 202 and converted to a rating.

For example, if the main decision facet "situation face" is set to X ₅ , one of the plurality of facet factors that constitute the situation face, the "family stock strength indicator" can be set to Y ₅₁ , and The score corresponding to Y ₅₁ is set to X ₅₁ . Since the strong indicator of the ethnic stocks is the percentage of the increase of the group stocks compared with the increase of the market, the processor 201 can obtain the increase of the ethnic group stocks and the market growth by the market open database 202, and calculate the group stocks. Inferior index Y _51, and further based on pre-stored in databases connected to the processor 201 or the processor of the group table built stocks index Y ₅₁ Inferior converted to a score of 203 X _51, the table can be For example, the following is shown:

When Y ₅₁ = group stocks strong indicators = group stocks rose, the broader market gains, Y ₅₁ is used to compare the rise and fall of a group of stocks compared to the strength of the broader market, including Y ₅₁ and X ₅₁ The corresponding relationship is explained as follows: When the group stocks increase by >0 and Y ₅₁ ≧1.2: the increase of the group stocks and the market growth are positive, that is, the ethnic stocks and the broader market are both rising, and the stocks of the ethnic groups are more successful. In the broader market, the corresponding X ₅₁ is 1 point; when the group stocks increase by >0 and 0.8≦Y ₅₁ <1.2: the increase of the group stocks and the market growth are positive, that is, the ethnic group stocks and the broader market are rising. And the increase of ethnic stocks is similar to the broader market, so the corresponding X ₅₁ is 0.5 points; when the ethnic stocks increase by >0 and Y ₅₁ <0.8: when 0≦Y ₅₁ <0.8 indicates that the stocks of the ethnic stocks are positive and The market's gains were positive, but the increase of ethnic stocks was weaker than the broader market. Therefore, it is recommended to wait and see. The corresponding X ₅₁ is 0 points. When Y ₅₁ <0, it means that the stocks of the ethnic groups are positive, but the market gains are negative. At this time, the high recovery may suffer downward price correction, hence recommended wait, the corresponding X ₅₁ is 0; when group Stocks rose <0 and Y ₅₁ <0.8: when 0 ≦ Y ₅₁ <0.8 represents a sector group is negative and increases gains were negative, but the group underperform stocks decline, it is suggested wait, the corresponding X ₅₁ is 0 points. When Y ₅₁ <0, it means that the increase of the group stocks is negative, but the market growth is positive. At this time, the status of the ethnic stocks is unknown. Therefore, it is recommended to wait and see, and the corresponding X ₅₁ is 0 points; When the group stocks rose by <0 and 0.8≦Y ₅₁ <1.2: the growth of the group stocks and the broader market were negative, that is, the ethnic stocks and the broader market were both down, but the decline of the ethnic stocks was similar to the broader market, so The corresponding X ₅₁ is -0.5 points; and when the group stocks increase by <0 and Y ₅₁ ≧1.2: the increase of the group stocks and the broader market are negative, that is, the ethnic stocks and the broader market are both falling, but the ethnic groups The decline of the stock is better than the decline of the broader market, so the corresponding X ₅₁ is -1 points.

The above score X ₅₁ is used to indicate the extent and importance of the event represented by the facet factor and will be substituted for the intensity equation (A).

For another example, the other facet factor of the plurality of facet factors constituting the main facet "face" X ₅ may be set to Y ₅₃ and the score corresponding to Y ₅₃ may be set. X ₅₃ . Please refer to Figure 3. Figure 3 shows the "Reverse Line" and "Morning Star" line types recommended by investors in the famous "Kalefield Warfare" textbook. The "Star" line type and the "Chuanzi Sanhe" line type that investors are advised to watch. After the invention obtains the K-line type of an investment project from the market disclosure database 202, the processor 201 compares the K-line type of the investment project with the winefield battle-type data stored in the database 203 connected to the processor. If the comparison K-stack 201 finds that the K-line type actually generated by the investment item matches the typical K-line type described in the "Wine Field Method", the processor 201 will follow the different typical K lines in the "Wine Field Method". For the buy, sell or wait-and-see suggestions corresponding to the line type, the score X ₅₃ corresponding to the facet factor "Jiutian Line Type Judgment" is obtained according to, for example, the following table:

In the above table, "Night Star", "Morning Star", "Counter Strike Line" and "Chuan Character Three Black" are typical K-line types described in "Wine Field Warfare". According to the "Wine Field Warfare Law", when the K-line type of the investment project is the same as "Night Star", it is recommended to sell; when the K-line type of the investment project is the same as "Morning Star" or "Reverse Line", it is recommended to buy; When the K-line type of the investment project is the same as "Chuanzi Sanhe", it is recommended to wait and see. So after comparing processor 201, if the K-line style of investment projects in line with "Evening Star", the processor 201 will correspond to the facets of factors, "Sakata line judge" X ₅₃ Y ₅₃ ratings points to -1; if If the K-line type of the investment project is in line with "Morning Star" or "Reverse Line", then X ₅₃ =1; if the K-line type of the investment project is in accordance with "Chuanzi Sanhe", then X ₅₃ =0, except for the line type exemplified in the above table. The "Winefield Warfare" also contains a number of different typical line types for comparison, and each has a corresponding buy, sell or wait-and-see advice for the processor 201 to score. The above score X ₅₃ is used to indicate the extent and importance of the event represented by the facet factor and will be substituted for the intensity equation (A). Similarly, the remaining facet factors can also be retrieved from the market open database 202, and then the corresponding scores can be obtained through the calculation process or the database comparison process and the look-up table method to replace the intensity equation (A). .

The pressure support surface may be composed of a plurality of facet factors, which may include a 20-day price-weighted average price (monthly average price) conditional judgment, a 60-day price-weighted average price (quarter average price) conditional judgment, and a 120-day price-weighted average Price (half-year average price) condition judgment, 240-day price-weighted average price (annual average price) condition judgment, pressure support strength index, each of which can be obtained by the market public database 202 and converted into a score.

For example, if the main control facet "pressure support surface" is set to X ₆ , one of the plurality of facet factors that constitute the pressure support surface may be the 20-day price-weighted average price (monthly average price). "Conditional judgment" is set to Y ₆₁ , and the score corresponding to Y ₆₁ is set to X ₆₁ . The price-weighted average price is the weighted average price obtained by averaging the price of an investment project weighted by the volume. For example, if a stock trades 10,000 shares at 20.5 yuan for a period of time and also trades 5,000 shares at 20.8 yuan, the weighted average price of the stock in this case is (20.5×10000+20.8×5000). ÷ (10000+5000) = 20.6 yuan. Calculate the weighted average price after a fixed time such as 20th, 60th, 120th or 240th, and you can get the weighted average price of the price on the 20th, 60th, 120th or 240th day. . Take the 20-day price-weighted average price as an example. Since the trading day of one month is about 20 days, the weighted average price of the 20-day price can also be called the monthly average price. Similarly, the weighted average price of the 60-day price is also It can be called the average price of the season. The weighted average price of the 120-day price can also be called the half-year average price, and the weighted average price of the 240-day price can also be called the annual average price. The processor 201 can obtain the monthly average price, the average daily price, the half-year average price, and the annual average price of the daily updated data from the market disclosure database 202, and draw the change into a monthly average, a daily average, and a half year. Moving average and annual average. Since these moving averages are drawn from long-term price changes, the changes are usually relatively gentle and can be used as a reference for the support and pressure surfaces of the price. The support surface refers to a lower limit and the pressure surface refers to an upper limit. If the price of the investment project, if the stock price rises above the monthly average, it means that the stock price has already risen to the pressure side, so the processor 201 can give the corresponding score X ₆₁ a positive score according to the built-in comparison table; otherwise, when the price of the investment project If the stock price falls below the monthly average, it indicates that the stock price has fallen below the support level, so the processor 201 can give a negative score to the corresponding score X ₆₁ according to the built-in comparison table. The comparison table can be as follows, for example:

The above score X ₆₁ is used to indicate the extent and importance of the event represented by the facet factor and will be substituted back to the intensity equation (A).

Factors other facets and a plurality of facets, for example, when the main decision factors facets "support face pressure" is set to X _6, the support surface may be composed of a pressure in the "pressure support strength index" to Y _65, and The score corresponding to Y ₆₅ is set to X ₆₅ . Generally speaking, the support strength calculation method is as follows: If today's closing price>yesterday's closing price

Then today's support strength = today's closing price - today's support point

The support point formula is: If today's lowest price < yesterday's closing price, today's support point = today's lowest price, if today's lowest price ≧ yesterday's closing price, then today's support point = yesterday's closing price; in addition, the general pressure intensity The calculation method is as follows: If today's closing price < yesterday's closing price

Then today's pressure intensity = today's pressure point - today's closing price

The pressure point formula: If today's highest price > yesterday's closing price, today's pressure point = today's highest price; if today's highest price ≦ yesterday's closing price, then today's pressure point = yesterday's closing price.

Therefore, after the daily closing, the processor 201 can obtain the support strength and the pressure intensity of a specific item according to the data of the market disclosure database 202, and give a corresponding score X ₆₅ according to, for example, the following table:

When the pressure intensity greater than the intensity of the support, increase the strength of strong, it is recommended to buy more, and may correspond to the rates Y ₆₅ X ₆₅ min to 1, whereas when the pressure is less than the strength of the supporting strength, the strength of strong decline, Therefore, Y ₆₅ can be set to -1 point. The above score X ₆₅ is used to indicate the extent and importance of the event represented by the facet factor and will be substituted for the intensity equation (A). Similarly, the remaining facet factors can also be retrieved from the market open database 202, and then the corresponding scores are obtained through the calculation process and the look-up table method to replace the intensity equation (A).

According to an embodiment of the present invention, the processor 201 weights the weighting factors W ₁ to W _{n for} a plurality of main decision surface planes X ₁ to X _n according to the initial intensity equation (A) for an investment item such as a specific stock. And corresponding to a plurality of facet factors, an initial buy signal strength I is calculated. When the buy signal strength I is greater than a user preset threshold buy signal strength I _th , the processor will send an inquiry message to the user's smart phone, stock machine and other handheld electronic equipment or personal computer to notify the user. Or the server decides whether to buy.

Compared with the prior art, each of the main decision surface planes X ₁ to X _n and their corresponding plurality of facet factors W ₁ to W _{n are} only determined by a preset condition as a condition for triggering the notification, except that the present invention The weighting factors X ₁ to X _n weight each of the main decision surfaces to indicate their influence, and further optimize the weighting factors W ₁ to W _n so that the intensity equation (A) calculates the obtained buy signal. Intensity I, after considering the optimization of previous transaction records, more accurately represents the degree of buyability of an investment project.

According to an embodiment of the present invention, the weighting factors W ₁ to W _{n are} optimized as described above, and the processor 201 may collect the previous transaction data for the time period corresponding to the main decision structure after a long period of time. The scores X ₁ to X _n and the Bayesian Optimization Algorithm (BOA), which has been commonly applied in the field of machine learning, is used to optimize the complex number in the intensity equation (A) Weight factors W ₁ to W _n . The previous transaction data may be the previous transaction data actually occurring in a long period of time, and the length of the longer period may be set by the user, for example, may be set to one week, that is, the processor 201 may analyze the week every other week. Previous transaction data, optimized by the Bayesian optimization procedure for the weighting factors W ₁ to W _n , and a set of optimized weighting factors W ₁ to W _n , according to which the intensity equation (A) is updated . After the above optimization, the weighting factor in the intensity equation (A) is replaced by the optimized weighting factor, and the strength equation (A) is thus optimized to an optimized intensity equation (A). ). Subsequent use of the purchased signal strength I calculated by the optimized intensity equation (A) can more accurately reflect good buying opportunities. According to an embodiment of the present invention, the step of performing the Bayesian optimization procedure on the weighting factors W ₁ to W _n may be as follows: Step 1: Define a starting strength equation (A) as , where the Lord decides the facet Where a is a positive integer from 1 to n, and the weighting factor W _a is set by the user to a preset value; Step 2: The processor 201 obtains the public information from the market public disclosure database 202, thereby obtaining the information in step 1 The master decides the facet X _a and obtains an initial buy signal strength I; step 3: the user specifies a longer period, such as one week or one month, to accumulate sufficient buy signal strength I value and The I values corresponding to the main decision facet X _a ; Step 4: Bayesian network (Baysian Network) each main decision facet X _a as a node to consider the interaction of each main decision surface X _a The conditional probability, taking the present invention as an example, the result to be obtained is a set of weighting factors W _a that maximize the probability of occurrence of the maximum I value corresponding to the best buying time point, where a is a positive integer of 1 to n Step 5: After substituting the previous transaction data accumulated in the third period in the segment into the Bayesian network of step 4, the processor 201 uses, for example, a Markov chain Monte Carlo algorithm. To analyze:

Among them: Markovbket is a mesh structure function describing causality, and each node is an event, which is used to link the causal relationship between events represented by each main decision face W _a ; Children are the result of an event, for example, Y _a belongs to the result event with the weight factor W _a as the cause; and parent is the event event of the event, which is used to determine the intensity corresponding to the maximum buy signal. The I value has one of the highest probability group weighting factors W _a , where a is a positive integer from 1 to n; Step 6: A new set of new samples is sampled by the processor 201 using the Markov chain Monte Carlo algorithm exemplified in step 5 the weighting factors Wa, then as a new set of weighting factors W _a, generations back to step 5. So iterate in a loop until it is impossible to find a set of weighting factors W _a that make the probability of the maximum I value corresponding to the best buying point appear, which is the termination condition, and a set of weighting factors W _{a is} a set of optimized weighting factors W ₁ to W _n . After the termination condition is reached, step 6 ends and proceeds to step 7.

Step 7: Substitute a group of updated weighting factors W _a and replace them with I= , that is, the intensity equation (A), used as the calculation of the buy signal strength I in the future.

The weighting factors W ₁ to W _n described above using the Markov chain Monte Carlo algorithm to perform a Bayesian optimization algorithm to obtain a set of optimizations are merely illustrative of the present invention and are not intended to limit the present invention. The scope of the algorithm can also be used to obtain a set of optimized weighting factors W ₁ to W _n , but all need to follow the Bayesian optimization procedure, that is, the optimized result is returned to the most The optimization algorithm, such an iteration loop, is repeated and optimized until a better result cannot be obtained. In this way, a set of optimized weighting factors W ₁ to W _n can be obtained, which can match the maximum buying strength I with the historical best buying point according to the previous transaction data. The weighting factors W ₁ to W _{n are} the final results of the sub-optimization, which are used to replace the intensity equation (A). After the completion of the Bayesian optimization procedure, the calculation of the buy signal strength I is used. The Bayesian optimization procedure is performed on the weighting factors W ₁ to W _n with the previous transaction data for a longer period of time as described above in order to make the intensity equation (A) more accurately reflect the real market change, so that The calculated buy signal strength I can correspond to a good buy point with a higher probability.

According to an embodiment of the present invention, the above-mentioned buy signal strength I can be used as a reference for the investment proportion, in addition to assisting the user or the server to determine whether to purchase an investment project. For example, according to the value of the purchase signal intensity I to determine the purchase amount, it can be _expressed as follows: if I _th2 >I≧I _th1 then C=C ₁ and C ₁ =α ₁ % ^*the average daily trading volume; if I _th3 > I ≧ I _th2 then C = C ₂ and C ₂ = α ₂ % ^* single-day average trading volume; and if I ≧ I _th3 then C = C ₃ and C ₃ = α ₃ % ^* single-day average trading volume, where: I is the buy signal strength; C is a purchase amount, which is a unit of measure that can be set by a user; I _th1 , I _th2 and I _th3 are respectively a buy signal strength; C ₁ , C ₂ and C ₃ is a quantity of funds, respectively, wherein C ₃ > C ₂ > C ₁ ; and α ₁ , α ₂ and α ₃ are respectively a user-definable ratio, and α ₃ > α ₂ > α ₁ .

According to an embodiment of the present invention, the above-mentioned purchase signal strength I can assist the user in judging whether to buy an investment item or not, and can assist the user in judging whether to sell an investment item. It can be seen from the above that when the value of the buy signal strength I of an investment project is larger, the more it is worth buying, the less obvious the value of the buy signal strength I of an investment project is as small (if I is negative and its The larger the absolute value, the more it is worth selling. When the processor 201 may be to buy signal intensity I is less than a threshold intensity sold _I thsell, notification to the user of handheld electronic device and / or a personal computer 204, notifying the user to decide whether to sell or server. Users can also set the stop loss or stop profit at the time of sale according to their needs. The setting conditions can be as follows: the percentage of single stop, where the user can set the target stop percentage for a single sale; Percentage of stop loss, where the user can set a target stop loss percentage for a single sell; k stop percentage, where the user can set a target stop percentage for k sells, and the number k can be set by the user Where k is a positive integer and k≧1; and a floating stop loss percentage, where the user can set a target stop loss percentage and the average closing price of one of the most recent j days as indicated by the simple moving average SMA(j) And the purchase cost of the previously purchased investment item is compared with each other to obtain a recommendation to sell, wherein j can be set by the user, j is a positive integer and j ≧ 1.

For example, if a condition "floating stop loss percentage" is set to S ₁ , then the nearest j-day simple moving average SMA(j) should be referred to, where the simple moving average SMA(j) is the most recent investment item. The first-line chart drawn continuously by the average price of the daily closing price, wherein the number of days j can be set by the user. The user can also set the c value of the investment item's price compared to the previous purchase price's stop loss percentage c%, where 0≦c≦100, that is to say (the price of the investment item ÷ the previous purchase price) When ≦(100-c)%, the c% stop loss point is reached. For example, if j=5 and c=20, when the buy signal strength I sells the strength I _thsell , and the price of the investment item has fallen below the moving average SMA (5) of the last 5 days, and If the price of the investment item has reached 20% stop loss compared to the purchase price, the processor 201 will issue a notification to the user's handheld electronic device and/or a personal computer 204 for the user or the server to decide whether to sell. Please refer to the following form:

Similarly, the remaining stop or stop condition for judging whether to sell or not may be obtained by the processor 201 from the market disclosure database 202, and then sent to the processor 201 for notification by the processor 201 or by means of a calculation process. The server decides whether to sell.

Please refer to Figure 4. FIG. 4 is a flowchart of a method according to the present invention. The method for reporting a strategy according to the present invention may be as follows: Step 401: The processor 201 has an initial intensity equation (A), and the initial intensity equation (A) A plurality of main decision surface planes X ₁ to X _n and a plurality of corresponding weight factors W ₁ to W _{n are included} ; Step 402: The processor 201 optimizes the plurality of weight factors according to the previous transaction data and the Bayesian optimization. W ₁ to W _n to generate an updated intensity equation (A); Step 403: The processor 201 provides a recommendation to adjust the portfolio based on the updated intensity equation (A). And after a period of time, iteratively performs step 402.

The "period" in step 403 can be set by the user, for example, one week or one month, and the previous transaction data in step 402 is the accumulated transaction data in this period.

In summary, compared with the prior art, the user can only be notified on a case-by-case basis when the preset event is sporadically present, and the present invention can not only comprehensively consider many main decision-making faces affecting the investment strategy, but further corresponding to the plurality of main decision-making faces. Determining the weighting factors of the facets, respectively weighting the importance and influence of the main decision facets and their respective facet factors on the strength of the buy signal, so as to propose to buy or sell the investment advice to the server or Users, and can further recommend the investment amount and proportion; in addition, the present invention can also calculate the investment according to the Bayesian optimization procedure commonly used in the machine learning field according to the previous transaction data actually occurring on the market. The plurality of weighting factors in the proposed intensity equation are optimized, so the intensity equation can be updated periodically or irregularly to maintain the correlation between the intensity equation and the real market and the accuracy of the calculation, thereby improving the previous In technology, only users can be notified when sporadic events occur, so they cannot integrate various investment information to provide investment advice and asset allocation. Shortcomings.

While the invention has been described above by way of example, it is not intended to limit the invention, and the invention may be variously modified and modified without departing from the spirit and scope of the invention.

401, 402, 403‧‧ steps

Claims (11)

A method of strategic trading, comprising: providing a starting intensity equation using a processor, the starting intensity equation comprising a plurality of main determining facets and a plurality of corresponding weighting factors; and using the processor to optimize the previous trading data A plurality of weighting factors are used to generate an updated intensity equation. The method of claim 1, further comprising: adjusting the portfolio according to the updated intensity equation. The method of claim 2, wherein adjusting the portfolio according to the updated intensity equation comprises: increasing or decreasing the proportion of an investment item based on the updated intensity equation. The method of claim 1, further comprising: providing a proposal to adjust the portfolio based on the updated intensity equation. The method of claim 4, wherein the suggesting to provide an adjusted investment portfolio based on the updated intensity equation comprises: providing a recommendation to increase or decrease the proportion of an investment project based on the updated intensity equation. The method of claim 4, wherein the providing the adjusted portfolio based on the updated intensity equation comprises providing a proposal to adjust the portfolio to a handheld electronic device and/or a personal computer based on the updated intensity equation. The method of claim 1, wherein each of the main decision facets comprises at least one facet factor. The method of claim 1, wherein the plurality of main decision facets comprises a price kinetic energy surface, a fundamental surface, a technical surface, a chip surface, a surface surface, and/or a pressure support surface. The method of claim 1, wherein the processor is used to optimize the plurality of weighting factors based on previous transaction data by using the processor according to previous transaction data and a Bayesian Optimization Algorithm; BOA) optimizes the plurality of weighting factors. The method of claim 1, wherein the processor is used to optimize the plurality of weighting factors based on prior transaction data by using the processor to optimize the plurality of weighting factors based on previous transaction data for a predetermined time period. The method of claim 1, wherein the processor is used to optimize the plurality of weighting factors based on prior transaction data by using the processor to optimize the plurality of weighting factors based on prior transaction data at a rest time.