TWI911511B - Trendline Method - Google Patents

Abstract

One method of trend line generation involves calculating and predicting the endpoints of existing trading chart areas to create trend lines. It can continuously generate new endpoints or eliminate old endpoints based on the latest real-time data (such as the latest prices and fluctuations of financial products), thereby continuously building future trend lines or applying them to any trend line (or any other line type, such as straight lines, curves, or broken lines) connected by endpoints.

Description

Trendline Method

This invention relates to a method for generating trend lines by predicting the endpoints of existing trading chart areas.

Traditional trend lines, being simply drawn as a single line, require strategy, human intervention, and programming for creation. Therefore, they can only serve as visual/passive alerts for traders and are not applicable to automated trading programs. Furthermore, current trend line offerings suffer from a lack of variety, or even absence of trend lines, in both computer and mobile device (phone/tablet) versions. Trend lines are difficult for traders to use; the current method fails to provide predicted values for subsequent trading times, requiring investors to use rulers or other tools to predict or calculate these values themselves, which is very inconvenient and beyond the capabilities of most investors; furthermore, the current trend lines can only be attached to a single time level and cannot simultaneously display trend lines for multiple time levels.

The applicant used the following search keywords, such as: ((stock market) OR (stock))(trendline)(endpoint), (stock market)(stock)(Trendline), (stock market)(stock)(Trendline)(trade), (~patent/TW201606679A), (stock market)(stock)(Trendline)(~patent/TWI721604B)(~patent/TW201606679A), (stock market)(stock)(Trendline)(~patent/TWI721604B)(~patent/TW201606679A)country:CN,TW; and reviewed 6600 patents through cross-search, selecting six prior cases, such as:

As in Case 1 above, it is a US patent for "Threshold trading method" (publication number: US20060265311A1), which is an investment method for the stock market that utilizes at least two trends. The steps include identifying the first trend based on the stock's recent past activity and determining the next longer trend; and executing a trade to follow the next longer trend. However, the case does not disclose how to simplify the number of trend lines by distinguishing between long, medium and short trends for comparison, continuously updating the output of each trend line, and simplifying the number of indicators for traders.

As in Case 2 above, it is a Republic of China patent for "a method for providing visual investment information on individual stock trend charts, a non-temporary computer-readable recording medium and information system" (Announcement No.: TWI721604B); it does not simply present markers, but draws the trend line of the entire traded commodity.

As in Case 3 above, it is a patent of the Republic of China, "Stock Information Advice System" (Announcement No.: TW201305953A). It does not draw trend lines based on buy and sell information, but rather calculates and connects trend lines based on endpoints.

Case 4 above is a Chinese patent for "Three-segment three-order mathematical model and its application in securities trading" (Announcement No.: CN1584928A). It uses the 3TM function for calculation, but it does not disclose the simple capture of point positions to identify endpoints, nor does it disclose the fact that trend lines do not need to be drawn by identifying high and low prices.

As mentioned in Case 5 above, it is a Chinese mainland patent for "Training Method of Data Trend Prediction Model and Prediction Method of Data Trend" (Announcement No.: CN112580800A), which is an algorithm for identifying line changes through image recognition, not trend line drawing.

Case 6 above is a Republic of China patent "Method for Indicating Trends in Securities Information" (Publication No.: TW201606679A). It is a trend line drawn within a fixed range, which is a manual processing method. It does not disclose that the trend is drawn simply by identifying future trend endpoints, nor does it disclose that it is not subject to range restrictions.

In addition to failing to disclose the aforementioned technologies, the previous case also failed to disclose the methods for establishing trend lines by connecting endpoints, endpoint finding algorithms, critical point algorithms, presenting future trend lines, and generating customized trend lines. Furthermore, it still lacks certain features in its products, which greatly reduces its practicality. This is an area that both businesses and consumers are eager to address.

To address the shortcomings of the existing technology, the main objective of this invention is to provide a trend line method including an endpoint finding algorithm, a critical point algorithm, a trend line algorithm, and a trend line drawing or output completion, in order to overcome the difficulties in the existing technology.

The secondary objective of this invention is to provide a trend line method that generates trend lines by predicting and calculating the endpoints of the current trading chart area. New endpoints can be continuously generated for the latest real-time data, or old endpoints can be eliminated, and future trend lines can be continuously built accordingly. Trend lines of different time levels can be applied to the window of the time level in effect.

Another objective of this invention is to provide a trend line method that effectively improves ease of use and establishes quantifiable trading risks.

The problem this invention aims to solve is that traditional/habitual trend lines are drawn as a single line. This drawing requires existing strategies, human intervention, and programming, thus it can only serve as a visual/passive reminder for traders and has no application in automated program trading. Furthermore, as mentioned in the aforementioned US patent "Threshold trading"... The method (Announcement No.: US20060265311A1) does not disclose the comparison of long, medium and short trends, but simplifies the number of trend lines by continuously updating each trend line and re-outputting it, thus simplifying the number of indicators for traders; similar to the Republic of China patent in the previous case 2, "Method for providing visual aids to investment information on individual stock trend charts, non-temporary computer-readable recording medium and information system" (Announcement No.: TWI721604B); it does not simply present Instead of marking, it draws the trend line of the entire traded commodity; for example, the Republic of China patent "Stock Information Advice Provision System" (Announcement No.: TW201305953A) in the previous case does not draw the trend line based on buy and sell information, but rather calculates and connects the trend lines based on endpoints; similarly, the Chinese mainland patent "Three-segment Three-order Mathematical Model and Its Application in Securities Trading" (Announcement No.: CN1584928A) in the previous case calculates the trend line using the 3TM function, without disclosing the simple capture of... Point identification endpoints do not disclose the need to draw trend lines by identifying high and low prices; for example, the Chinese mainland patent No. 5, "Training Method for Data Trend Prediction Model and Method for Predicting Data Trends" (Announcement No.: CN112580800A), uses an algorithm to identify line changes through image recognition, not to draw trend lines; similarly, the Republic of China patent No. 6, "Method for Indicating Trends in Securities Information" (Announcement No.: TW201606679A), uses a fixed range... The trend lines drawn internally are a manual processing method. They do not disclose how the trend is drawn simply by identifying future trend endpoints, nor do they disclose that they are not subject to range restrictions. In addition to not disclosing the aforementioned technologies, the previous case also did not disclose the methods for establishing trend lines by connecting endpoints, endpoint finding algorithms, critical point algorithms, presenting future trend lines, and generating customized trend lines. Furthermore, these shortcomings in the product significantly reduce its practicality, which is an area that both businesses and consumers are eager to address.

To achieve the above objectives, the present invention provides a trend line method, the steps of which include: an endpoint finding algorithm, which provides at least one endpoint data; the endpoint finding algorithm refers to a method for establishing high/low endpoints; it obtains endpoints by sorting the data of each point, initiating the calculation based on the highest/lowest value, distinguishing the left and right intervals based on the highest/lowest value, and continuously indexing the descending/ascending sort of each time point; A critical point algorithm is connected to the endpoint finding algorithm. The endpoint data obtained by the endpoint finding algorithm is filtered by the critical point algorithm to select at least one critical point data. The critical point data includes at least one vertex and at least one inversion point. The critical point algorithm refers to finding the correlation between the high and low values of the endpoint data and determining whether each endpoint is a vertex or an inversion point according to the time sequence time point value, so as to complete the calculation operation of defining critical points such as vertices and inversion points. A trend line algorithm is connected to the critical point algorithm. The trend line algorithm obtains the data of the endpoints and critical points from the endpoint search algorithm and the critical point algorithm, and connects the critical points to form a trend line to complete and output the trend line. A trend line drawing or output completion is completed by connecting the trend line algorithm and completing the drawing or output of the trend line. The trend line output includes, but is not limited to, a trend line data table, a trend line matrix data table, a trend line chart, etc. The trend line is generated by predicting and calculating the endpoints of the current trading chart area. New endpoints can be continuously generated or old endpoints can be eliminated based on the latest real-time data. Future trend lines can be continuously built in this way. Trend lines of different time levels can be applied to the window of the time level in effect.

In this invention, the endpoint data and the critical point data can be endpoint, vertex, and inversion point data in the Critical Point Table (CPT).

The trend line algorithm of this invention can form trend lines including, but not limited to, a vertex-to-vertex form, a vertex-to-reversal form, an end-to-end form, a slope-based filtering form, and a distance-based filtering form.

Specifically, the peak-to-peak connection method of this invention connects peaks in the Critical Point Data Table (CPT) to form a trend line that can show both upward and downward trends; the peak-to-reversal connection method connects peaks in the CPT to form a trend line that can show both upward and downward trends; the endpoint-to-endpoint connection method connects peaks in the CPT to form a trend line that can show both upward and downward trends. The trend line in the trend line data table (TT) consists of a vertex, a reversal point, and an endpoint connected to the endpoints. The slope selection method is to select the trend lines in the trend line data table (TT) based on their slopes being close to a certain slope or having a large difference. The interval selection method is to select the trend lines in the trend line data table (TT) based on their endpoint values being greater than or less than a certain interval.

In this invention, the completed trend line output can be output to the program system interface of the trading product. It can display the subsequent trend line prediction and endpoint values through a subsequent trend line prediction area, a trend line control area, and an endpoint prompt control area as the trading time progresses.

The trend line method of this invention can be applied to single or multiple preceding market conditions. Through the setting operation/adjustment of trend lines, it provides a method to set multiple endpoints to form trend lines.

In this invention, the predecessor market refers to the market results of the country that opened for trading earlier. The country that opened for trading earlier is the predecessor market. For single or multiple predecessor market data, the invention provides a method to set multiple endpoints to form a trend line through the setting operation/adjustment of trend lines.

The endpoint finding algorithm, critical point algorithm, and trend line algorithm of this invention establish endpoints, critical points, and trend lines by including, but not limited to, node data, hash data, and a critical point data source. The node data refers to the relevant records of the main nodes, including, but not limited to, time point index, time point, K-bar's highest value, lowest value, closing value, opening value, candidate record value, left high point record, right high point record, left low point record, and right low point record. The hash data refers to the data obtained through a hash function (Hash). The function (HF) is used to calculate the position corresponding to a key and a value; the critical point data refers to the points within the trend line endpoints, including but not limited to high points, peaks, and reversal points.

The completed trend line drawing or output result data table of the present invention includes, but is not limited to, a trend line data and a trend line matrix data; the trend line data refers to the trend formed by combining data from regions divided by the critical point, the vertex, and the reversal point. The trend line data mentioned above includes, but is not limited to, the types of trend lines from upper peak to lower peak, lower peak to upper peak, peak to reversal point, reversal point to peak, endpoint to peak, endpoint to endpoint, etc.; the trend line matrix data can combine or discard the trend lines in the trend line data.

Compared to the effectiveness of prior art, the trend line method of this invention includes the endpoint finding algorithm, the critical point algorithm, the trend line algorithm, and the completion of trend line drawing or output; thereby achieving the formation of a trend line through the prediction calculation of the endpoints of the current trading chart area; it can continuously generate new endpoints or eliminate old endpoints based on the latest real-time data, and continuously build future trend lines accordingly. Trend lines of different time levels can be applied to the window of the time level in effect; effectively improving its ease of use and making trading risks quantifiable, which can greatly expand the industry's utilization and is novel and progressive.

The present invention is described in detail below with reference to the accompanying drawings and in the form of embodiments. The drawings used in this document are for illustrative purposes only and are not necessarily the actual scale and precise configuration of the present invention in practice. Therefore, the scale and configuration of the accompanying drawings should not limit the scope of the present invention in actual practice.

Please refer to Figures 1 and 2 (A)(B)(C), 3 (A)(B), 4, 5, 6 (A)(B)(C)(D)(E), 7, 8, 9, 10, and 11, which are the block diagram, node data table (NT), node hodgepodge table (HT), critical point data table (CPT), and invention trend chart of this invention. The invention includes: a trendline data table (TT), a trendline matrix data table (TMT), a flowchart of the endpoint finding algorithm, a flowchart of the critical point algorithm, implementation diagrams of the trendline algorithm for connecting vertices, connecting vertices and reversal points, connecting endpoints to endpoints, using slope filtering, and using interval filtering, and a method for presenting future trendlines for traded products. The following diagrams illustrate the functionalities/interface implementation of the method for presenting future trend lines for trading products, including: the trend line control area and endpoint indication control area; the implementation of the method for presenting future trend lines for trading products using a chart pattern to display upward or downward trend charts; and the method for presenting future trend lines for trading products using a chart pattern. The following are examples of implementation diagrams: an example of a horizontally symmetrical trend chart using Jiang Bo's diagram for the function/interface of an easy-to-use product application; an example of an application of a computer-based operation/adjustment method for customizing trend lines; and, in a preferred embodiment, the trend line method of this invention includes the following steps: an endpoint finding algorithm 1, a critical point algorithm 2, a trend line algorithm 3, and a trend line drawing or output completion 4.

The aforementioned endpoint finding algorithm 1 (as shown in the first figure) provides the acquisition of at least one endpoint data 11; (the endpoint finding algorithm 1) refers to a method for establishing high/low endpoints; it obtains endpoints by sorting the data of each point, initiating calculations based on the highest/lowest value, using the highest/lowest value as a reference to distinguish the left and right intervals, and continuously indexing the descending/ascending sort of each time point to complete the endpoint acquisition. The endpoint data 11 can be endpoint, vertex, and inversion point data in the Critical Point Table (CPT); that is, the present invention establishes a trend line by: first finding endpoints... Algorithm 1 obtains endpoint data 11, then uses endpoint data 11 to filter its internal critical points (vertices, inversion points) using Critical Point Algorithm 2. Finally, using Trendline Algorithm 3, the trendline is drawn or output as completion 4. (The output includes, but is not limited to, a trendline data table, a trendline matrix data table, a trendline chart, etc.); (Please refer to Figure 4, which is a flowchart of the endpoint finding algorithm of this invention. The process is as follows: First, the endpoint finding method begins (processing); then, a data storage array is designed, including an endpoint candidate record area (default is 0, all...) (This is Hou's explanation) For example, Crec[X]=0, x=1~z, there are z time points (processing); then sort the data in descending order according to the highest value H (H->L) (processing); then, left-side processing=1, for example LP=1, right-side processing=1, for example RP=1; then, find the time point index of the highest value in the interval=n, which can be found by going down from SN. When Crec[n]=0, set the candidate record=1, for example, when Crec[n]=1, set the left high point=1, for example, when LP=1, then LH[n]=1, set the right high point=1. Point = 1, for example, when RP = 1, then RH[n] = 1, perform the following left and right candidate elimination processing (processing); then through judgment, when the interval ends, it ends, otherwise the left and right processing is performed, and if the judgment is not the end, the (left) and (right) processing is performed. The left side: record the candidates that will not be high points on the left side as 1, and in order: find the low points in order (i.e., there is a reversal), and record the candidates of these points as 1, for example, Crec[HT[x]] = 1, x = n-1~n-r time points; then through the left interval (Excluding candidates with a record of 1), continue the same calculation method, e.g., LP=1, RP=0; then, when there is no interval to be found, end and return; where, (left interval (excluding candidates with a record of 1), continue the same calculation method, e.g., LP=1, RP=0) and return to the previous step (find the time point index of the highest interval = n, which can be found by going down from SN, when Crec[n]=0, set the candidate record = 1, e.g., Crec[n]=1, set the left high point = 1, e.g., when LP=1, then LH[n] =1, set the right high point =1, for example, when RP=1, then RH[n]=1, perform the following left and right candidate elimination process); right side: record the candidates that will not be high points on the right side as 1, and in order: find the low points in sequence (i.e., there is a reversal), record the candidates of these points as 1, for example, Crec[HT[x]]=1, x=n-1~n-s time points; then, for the right interval (not including those with candidate records =1), continue the same calculation method, for example, LP=0, RP=1; then, when there is no interval to be found End and return; among them, (the right interval (not including candidates with record = 1), continue the same calculation method, e.g., LP = 0, RP = 1) and return to the previous step (find the time point index of the highest in the interval = n, which can be found by going down from SN, when Crec[n] = 0, set the candidate record = 1, e.g., Crec[n] = 1, set the left high point = 1, e.g., when LP = 1, then LH[n] = 1, set the right high point = 1, e.g., when RP = 1, then RH[n] = 1, and perform the following left and right candidate elimination processing); End Candidate records > 0 are all endpoints. The implementation in Figure 4 illustrates this by using the method of capturing high endpoints from the data table. First, sort the data of each point, and start the calculation with the highest value. List each value as a candidate record (as shown in the flow below, Crec, the record of the nth point is Crec[n]=0, and after processing to the mth time, it is Crec[n]=m). If it is a high endpoint record, it is 1. The final statistical record > 0 is all endpoints. In the implementation of capturing low endpoints, it is only necessary to convert the high endpoint capture in the implementation flow to low endpoint capture.

The aforementioned critical point algorithm 2 (as shown in the first figure) is connected to the endpoint finding algorithm 1. The endpoint data 11 obtained through the endpoint finding algorithm 1 is filtered by the critical point algorithm 2 to select at least one critical point 21. The critical point 21 data includes at least one vertex and at least one inversion point. The critical point algorithm 2 refers to finding the correlation between high and low values within the endpoint data 11 and determining whether each endpoint is a vertex or inversion point based on the time-series time point values, thereby completing the calculation operation to define critical points such as vertices and inversion points. The critical point 21 data can be endpoint, vertex, and inversion point data in the Critical Point Table (CPT). For vertex, inversion, and other critical points... The method for determining a point is as follows: Vertex: the highest point, and the one whose value is higher than its left and right adjacent endpoints; Reversal point: the one whose value is higher than its left and right adjacent endpoints. Regarding the actual implementation of the critical point algorithm, taking the extraction of high points from a data table as an example: the left high point record (LH) and right high point record (RH) in the previously obtained node data (NT) can be used to determine whether it is a vertex or a reversal point based on the aforementioned time sequence time point values. Please refer to Figure 5 for the flowchart of the critical point algorithm of this invention. The process is as follows: First, the calculation method for critical points (endpoints - vertex points - reversal points) begins; then, the values obtained through the endpoint search calculation method are... The endpoint data table NT (taking finding the high point as an example, LH and RH in NT are recorded) is filled into the critical point (endpoint-node-reversal point) data table CPT; then, the data is sorted in ascending order according to the time point index (the smaller time point comes first); then, the first point and the last point are endpoints CP[1]=0, CP[n]=0, where n is the number of endpoints; then, from the second time point to the last point, the data is checked sequentially, for example, the i-th point; then, if the values of the left and right adjacent points are higher, then it is the reversal point CP[i]=1; otherwise, it is checked again (if the values of the left and right adjacent points are lower), then it is the vertex CP[i]=1; otherwise, it proceeds (not yet concluded). If the condition is "continue if not yet finished," then return to the previous step (checking sequentially from the second time point to the last previous point, for example, the i-th point); and if the condition is "continue if not yet finished," then end the process and proceed to "end, output the critical point (endpoint-vertex-reversal point) data table"; and the aforementioned judgment... (If the value of the left and right adjacent nodes is higher, then if it is (the inversion point CP[i]=1), then proceed to (if it has not ended, continue; otherwise, end); and as mentioned above (if the value of the left and right adjacent nodes is lower, then if it is (the vertex CP[i]=1), then proceed to (if it has not ended, continue; otherwise, end).

The aforementioned trend line algorithm 3 (as shown in the first figure) is connected to the critical point algorithm 2. The trend line algorithm 3 obtains the data of the endpoint 11 and the critical point 21 from the aforementioned endpoint finding algorithm 1 and the critical point algorithm 2, and connects each critical point 21 to form a trend line for completion and output. The trend line forms that the trend line algorithm 3 can form include, but are not limited to, a vertex-to-vertex form 31, a vertex-to-reversal point form 32, an endpoint-to-endpoint form 33, a slope filtering form 34, and a distance filtering form. Select Form 35; This peak-to-peak connection form connects peaks in the Critical Point Data Table (CPT) to form a trend line that can show both upward and downward trends; This peak-to-reversal connection form connects peaks in the CPT to form a trend line that can show both upward and downward trends; This endpoint-to-endpoint form connects endpoints in the CPT that contain both peaks and reversals to form a trend line; This slope filtering form filters trend line data... The trend lines in the trend table (TT) are selected based on their slopes being close to or significantly different from a given slope. The interval selection method involves selecting trend lines from the trend table (TT) whose endpoint values are greater than or less than a certain interval. The endpoint and critical point (vertices, inversion points) data obtained from the aforementioned endpoint finding algorithm 1 and critical point algorithm 2 (the data can be the endpoint, vertex, and inversion point data from the critical point data table (CPT) generated in the aforementioned embodiment) are then... The connection of each critical point 21 forms the trend line and output (as mentioned above: trend line data table (TT)); the trend line algorithm 3 can form trend lines in the form of CPT table and TT table generated in the aforementioned embodiment, which can be the following, including but not limited to, the following embodiments of the present invention formed by trend line algorithm in the form of connecting vertices with vertices, connecting vertices and inversion points, connecting endpoints to endpoints, slope filtering, and spacing filtering, as shown in Figure 6 (A), (B), (C), (D), and (E);

1. Peak-to-peak connection method: (Please refer to Figure 6 (A)) Connect the peaks in the CPT table to form a trend line. This can be shown as an upward or downward trend. The endpoints (high points) in the time series shown in the figure are, for example: time point tn1, high point value Hn1.

2. Connection of Peaks and Reversal Points: (Refer to Figure 6 (B)) Connect the peaks and reversal points in the CPT table to form a trend line. This can be shown as an upward and downward trend; the endpoints (high points) in the time series shown in the figure are, for example: time point tn1, high point value Hn1.

3. End-to-end form: (Please refer to Figure 6 (C)) Connect the endpoints in the CPT table to form a trend line (including the apex and the reversal point); the implementation method for finding the high point is as follows: (1) Use the apex and the reversal point as the dividing line, (2) Connect the endpoints in the interval between the dividing lines to form a trend line; where the endpoints in the time sequence shown in the figure are examples (high points); Example: time point tn1, high point value Hn1.

4. Slope Filtering Method: (Please refer to Figure 6 (D) for reference). The trend lines in the TT table are filtered out based on their slope (e.g., 5 degrees) when they are close to or differ too much from a certain slope. The endpoints (high points) in the time series shown in the figure are: for example, time point tn1 and high point value Hn1.

5. Spacing Filtering Method: (Please refer to Figure 6 (E)) The trend lines in the TT table are filtered out when their endpoint values are greater than or less than a certain spacing (e.g., a spacing of 0 indicates that the trend lines are touching each other). The endpoints (high points) in the time sequence shown in the figure are: for example, time point tn1, high point value Hn1.

The aforementioned trend line drawing or output completion 4 (as shown in the first figure) connects to the trend line algorithm 3 and completes the trend line drawing or output. The trend line output includes, but is not limited to, a trend line data table, a trend line matrix data table, and a trend line chart. The trend line output completion of the trend line drawing or output completion 4 can be output to the program system interface of the trading product, which can be accessed through a subsequent trend line prediction area. It features a trendline control area and an endpoint indicator control area, which change over time and can be moved to display subsequent trendline predictions and endpoint values. It generates trendlines based on the predicted endpoints in the current trading chart area and continuously generates new endpoints or eliminates old endpoints based on the latest real-time data, thereby continuously building future trendlines. Trendlines of different time levels can be applied to the window of the current time level.

This invention provides a trend line method that can handle single or multiple predecessor market data. Through trend line setting and adjustment, it offers a method for forming trend lines by setting multiple endpoints. The predecessor market refers to the market results from the country where the market opened for trading earlier; this country is the predecessor market. The endpoint finding algorithm 1, the critical point algorithm 2, and the trend line algorithm 3 are derived by including, but not limited to, one node data point and one hash (…). The hash data, a source of critical point data, is used to establish endpoints, critical points, and trend lines; where the node data refers to the relevant records of each major node, including but not limited to time point index, time point, K-line high value, low value, closing value, opening value, candidate record value, left high point record, right high point record, left low point record, right low point record, etc.; the hash data refers to the data generated by the hash function (Hash). The function (HF) calculates the position corresponding to a key and value; the critical point data refers to the points within the trend line endpoints, including but not limited to high points, peaks, and reversal points; the peak refers to the highest point, where the values of the adjacent endpoints to its left and right are lower, and the reversal point refers to the adjacent endpoints to its left and right, where the values are higher; the aforementioned data can be obtained from the data tables shown in Figure 2 (A), (B), and (C) below as the source of the trend line calculation: please refer to Figure 2 (A), which is the node data table 51 (NODES) of data table 5. Table 5 (abbreviated as NT) includes, but is not limited to, serial number, time point index, time point, highest value, lowest value, opening value, closing value, candidate record value, left high point record, right high point record, left low point record, and right low point record; please refer to Figure 2 (B), which is Table 5's node miscellaneous table 52 (HASH TABLE) (abbreviated as HT), whose items include, but are not limited to, time point index and serial number; please refer to Figure 2 (C), which is Table 5's critical point table 53 (CRITICAL POINT TBALE) (abbreviated as CPT), whose items include, but are not limited to, serial number, time point index, and critical point (endpoint-node-reversal point) record.

The present invention provides a data table 6 representing the completed trend line drawing or output result 4, which includes, but is not limited to, a trend line data 610 and a trend line matrix data 620. The trend line data 610 refers to the data where the trend line is generated by combining data from regions divided by the vertices and reversal points within the critical points. The trend line data 610 includes… Includes, but is not limited to, the types of trend lines from top vertex to bottom vertex, bottom vertex to top vertex, vertex to inversion point, inversion point to vertex, endpoint to vertex, endpoint to endpoint, etc.; the trend line matrix data 620 can merge or discard the trend lines in the trend line data (e.g., only retain the trend lines within a specific slope range (e.g., trend lines with a slope of 5 degrees and within ±1 degree).

The aforementioned trend line data 610 and trend line matrix data 620 can be output as trend line data through the following data table 6: (Please refer to Figure 3 (A), which is the trend line data table of this invention; Trend Line Data Table 61 (TRADELINE) The table (abbreviated as TT) includes, but is not limited to, serial number, type, start time index, and end time index. The type includes, but is not limited to, 1. from top vertex to bottom vertex, 2. from bottom vertex to top vertex, 3. from vertex to inversion point, 4. from inversion point to vertex, 5. from vertex to vertex, and 6. from vertex to vertex. (Please refer to Figure 3 (B), which is the trend line matrix table of this invention. The trend line matrix table (TRADELINE MATRIX TABLE) 62 (abbreviated as TMT) includes, but is not limited to, time point index. The time point index includes, but is not limited to, vertex 1, vertex 2, vertex 3, vertex 4, vertex 5, vertex 6, vertex 7, and vertex 8.)

Further applications of this invention: (Please refer to Figures 7, 8 (A) and (B), 9, 10, and 11, which are functional/interface implementation diagrams of the method for presenting future trend lines of trading products according to this invention in trading product applications; functional/interface implementation diagrams of the trend line control area and endpoint prompt control area of the method for presenting future trend lines of trading products according to this invention; and the trading platform of this invention.) The following diagrams illustrate the functions/interfaces of the method for presenting future trend lines for trading products, including: an example of implementing an upward or downward trend chart using the Jiangbo chart; an example of implementing a horizontally symmetrical trend chart using the Jiangbo chart; and an example of implementing a custom trend line generation application using a computer-like operation/adjustment method.

I. Method for presenting future trend lines of trading products: (as shown in Figure 7); The trend line output completion of the present invention is output to the program system interface of the trading product. It can be displayed through a subsequent trend line prediction area, a trend line control area, and an endpoint prompt control area, which can move and display the subsequent trend line prediction and endpoint values as the trading time progresses; that is, the present invention outputs the trend line completion to the program or system interface of the trading product. It can be displayed through a subsequent trend line prediction area, a trend line control area, and an endpoint prompt control area, which can move and display the subsequent trend line prediction and endpoint values as the trading time progresses. It can handle different time levels (e.g., 1 minute, 1 hour...4 hours, day, week, month, unlimited) and can be applied to automated trading (program trading). Its functions/interface implementation for trading instruments (as shown in Figure 7) includes: a time-based trend line locking area (which can overlap), a current trading chart area, a future trend line prediction area, a trend line control area, and a current... The numerical axis and endpoint control area, in this creation, the seventh chart serves to display trend lines at different timeframes. This embodiment displays the 1-hour trend line and the daily trend line. In the current trading chart, the trend line displayed at the 1-hour timeframe level, as shown, represents the low points of each wave in an upward trend within 1 hour. Connecting these low points forms the 1-hour upward trend... The trend line is formed by connecting the highest point at the end of an uptrend on the 1-hour chart to the highest point of the 1-hour timeframe on the daily chart, thus creating the daily downtrend line (as shown by the dotted line at the top of the chart). The highest reversal point of the current uptrend on the 1-hour timeframe is then connected to the highest point of the subsequent 1-hour timeframe. The downtrend line is formed by connecting the highest points of the rebounds. The aforementioned trend line control area can be adjusted using, but is not limited to, indicator points (e.g., candlestick axis numbers), time points (time points at the time level of the action, such as 18:00 on 2022/01/07 for a 20-minute candlestick), and numerical values (e.g., the highest/lowest value of the candlestick). Another mentioned endpoint indication control area, used to provide the trend line, allows for setting the endpoints accessible for future transactions. This can be established using, but is not limited to, indicator points, time points, estimated values, and alert values. It can also be used to indicate single or multiple trend lines or for other purposes. It also includes trend line addition/deletion operations. Example of the function/interface of the trend line control area and the endpoint prompt control area (as shown in Figures (A) and (B)). In Figure (A), the trend line control area 71 of control area 7 includes, but is not limited to, trend line, use, delete, indicator, time, value, and endpoint; and the endpoint prompt control area 72 of control area 7 of Figure (B) includes, but is not limited to, adding trend line, deleting trend line, trend line, endpoint, indicator, time, estimated value, and alarm value.

The implementation examples of the functions/interfaces of trading commodity applications can also be implemented using the Jiangbo Chart (i.e., the price-volume relationship chart 8) as an example (as shown in Figures 9 and 10):

1. Upward or Downward Trend Chart 81 provides trend lines to predict upward/downward trends. Implementation example is shown below (as shown in Figure 9), which displays the current trading chart area, the subsequent trend line prediction area, the trend line control area, the current value axis, and the endpoint indicator control area.

2. Horizontal Symmetry Trend Chart 82 can be used to calculate/locate horizontal trend lines, providing estimated highs or lows. Implementation example (as shown in Figure 10). This chart displays the current trading chart area, the horizontal symmetry trend line, the predicted high, the subsequent trend line prediction area, the predicted low, the trend line control area, the current value axis, and the endpoint indication control area.

II. Application of Customized Trend Line Generation: (As shown in Figure 11, this is an example of the application of the customized trend line generation of this invention in the operation/adjustment method of a type of computer)

This invention provides a trend line method that can handle single or multiple predecessor market data. Through the setting and adjustment of trend lines, it provides a method to form trend lines by setting multiple endpoints. The predecessor market refers to the market results of the country that opened for trading earlier. That market is the predecessor market. In other words, the operation/adjustment interface 91 of the computer in the eleventh figure, which is a custom trend line application, includes, but is not limited to, formulas, historical records, predecessor markets, strategies, successor markets, time frame, trend lines & endpoint area.

This invention provides a trend line method that can handle single or multiple preceding markets (meaning, based on the market results of the country where the market opened first, that country is considered the preceding market. For example, in the United States, the preceding market can include, but is not limited to, Japan, Taiwan, Hong Kong, etc.). Through the setting and adjustment of trend lines, it provides a method for setting multiple endpoints to form trend lines. This application is used in a type of computer... The operation/adjustment method is an example (as shown in Figure 11); Formula: refers to the algorithm for generating trend lines, such as the method of the core technology mentioned above (in this example: Taiwan Futures: High-Low Time Average Method); Historical Records: extracts previously used/set settings records; Successor Market: refers to the option of using the successor market (in this example: Taiwan Futures); Strategy: refers to the introduction of trading strategies ( In this embodiment, it is: High and Low Time Averaging Method); Follower Market: refers to the follower market option, i.e., the target market used for the trend line/strategy (in this embodiment, it is: US Small Dow Jones); Time Scale (in this embodiment, it is: 20 minutes): is the time adjustment indicator; Calculator Button Block: establishes the operation means used for the strategy/formula; the user can press the "=" button to correspond to this... The aforementioned core technology generates endpoint and trend line data, which can then be output to the traded product via the "Apply" button. The program or system interface of this type of computer's operation/adjustment interface 91 further includes parameter settings for applying the trend line in the trend line & endpoint area on the right side of the computer's key block. For example, the items for endpoints #1 and #2 include indicators, time, estimated values, and alarm values.

Comparison and advantages of this invention with previous inventions:

Traditional trend lines, being simply drawn as a single line, require existing strategies, human intervention, and programming. Therefore, they can only serve as visual/passive reminders for traders and are not applicable to automated trading programs. This invention calculates actual endpoint data and then connects them. This endpoint data can be output for use by automated trading programs. Furthermore, the endpoint data includes time information, allowing automated trading programs to adjust their strategies in response to changes in trading time (e.g., defining order placement/deletion points, adjusting whether existing orders are retained/deactivated, etc.), automating human trading indicators and establishing quantifiable trading risks.

This invention provides a trend line method that generates trend lines by predicting the endpoints of existing trading chart areas. It can continuously generate new endpoints or eliminate old endpoints based on the latest real-time data (such as the latest prices and fluctuations of financial instruments), thus continuously building future trend lines. Therefore, this invention can construct trend lines corresponding to any dataset (such as various data tables, original trend line data, etc.). It can be applied to financial instrument charts with different timeframes (such as candlestick charts, moving average charts, etc.) and can be used to calculate future time predictions for financial instrument trading, achieving a more accurate prediction of trading strategies.

The core technology of this invention is to predict, control, and indicate the future endpoints of trend lines. The endpoints of trend lines can be adjusted through the interface; trend lines can be formed by setting endpoints, and the operation of the previous method can be incorporated into transaction monitoring (the traditional method is to draw trend lines manually or generate them by calculation); the trend line and endpoint calculator of this invention can generate endpoints and apply them to the method of this invention. Trend lines of different time levels can be applied to the window of the time level in effect; it can provide applications for automatic trading (program trading).

Features of this invention:

1. Feature 1: Establishing trend lines by connecting endpoints: New endpoints can be generated or old endpoints can be eliminated by continuously identifying them, and trend lines can be established by connecting the endpoints.

2. Feature 2, Endpoint Finding Algorithm: By sorting the data of each point, the algorithm is initiated based on the highest/lowest value. The left and right intervals are distinguished based on the highest/lowest value. The algorithm continuously indexes each time point in descending/ascending sorting to obtain the endpoint.

3. Feature 3, Critical Point Algorithm: Based on the endpoint data, find the correlation between high and low values, and determine whether each endpoint is a vertex or inversion point according to the time sequence time point value.

4. Feature 4: Presenting future trend lines: Through subsequent trend line prediction, trend line control, and endpoint prompt control, it progresses with the trading time and can dynamically display subsequent trend line predictions and endpoint values.

5. Feature 5: Custom trend line generation: For the preceding market, the trend line can be set and adjusted in a way that allows for setting multiple endpoints to form a trend line, such as using a computer.

The trend line method of this invention includes an endpoint finding algorithm 1, a critical point algorithm 2, a trend line algorithm 3, and a trend line drawing or output completion 4; thereby achieving the formation of a trend line through the prediction calculation of endpoints in the current trading chart area; it can continuously generate new endpoints or eliminate old endpoints based on the latest real-time data, and continuously build future trend lines accordingly. Trend lines of different time levels can be applied to the window of the time level in effect; effectively improving its ease of use and making trading risks quantifiable, which can greatly expand the industry's utilization and has novelty and progress.

In summary, this invention has indeed achieved the desired improved effect by breaking through the previous technology, and it is not something that those familiar with the technology could easily conceive of. Furthermore, this invention has not been disclosed before this application, and its novelty and progressiveness clearly meet the requirements for an invention patent application. Therefore, this invention application is filed in accordance with the law, and we respectfully request your authority to approve this invention patent application in order to encourage invention. We are deeply grateful for your assistance.

The embodiments described above are merely for illustrating the technical ideas and features of this invention. Their purpose is to enable those skilled in this art to understand the content of this invention and implement it accordingly. They should not be used to limit the scope of the patent of this invention. All equivalent changes or modifications made in accordance with the spirit of this invention should still be covered within the scope of the patent of this invention.

1: Endpoint Finding Algorithm 11: Endpoint Data 2: Critical Point Algorithm 21: Critical Point 3: Trend Line Algorithm 31: Vertex-to-Vertex Connection 32: Vertex-to-Reversal Connection 33: Endpoint-to-Endpoint Connection 34: Slope Filtering 35: Spacing Filtering 4: Trend Line Drawing or Output Complete 5: Data Tables 51: Node Data Table 52: Node Data Table 53: Critical Point Data Table 6: Data Tables 61: Trend Line Data Table 610: Trend Line Data 62: Trend Line Matrix Data Table 620: Trend Line Matrix Data 7: Control Region 70: Functions/Interface of Trading Instruments Application 71: Trend Line Control Area 72: Endpoint Indicator Control Area 8: Real-time Price-Volume Relationship Chart 81: Upward or Downward Trend Chart 82: Horizontally Symmetrical Trend Chart 9: Custom Trend Line Generation Application 91: Computer-like Operation/Adjustment Interface

Figure 1: Block diagram of the invention. Figure 2 (A), (B), (C): Node table (NT), Node Hash table (HT), Critical Point table (CPT) of the invention's data table. Figure 3 (A), (B): Trendline table (TT), Trendline Matrix table (TMT) of the invention's data table. Figure 4: Flowchart of the endpoint finding algorithm of the invention. Figure 5: Flowchart of the critical point algorithm of the invention. Figure 6 (A), (B), (C), (D), (E): Examples of the vertex-to-vertex connection, vertex-to-reverse-point connection, endpoint-to-endpoint connection, slope filtering, and spacing filtering forms generated by the trendline algorithm of the invention. Figure 7: An example of the function/interface implementation of the method for presenting future trend lines for trading products according to the present invention in the application of trading products. Figure 8 (A) (B): An example of the function/interface implementation of the trend line control area and endpoint indication control area of the method for presenting future trend lines for trading products according to the present invention. Figure 9: An example of the function/interface implementation of the method for presenting future trend lines for trading products according to the present invention in the application of trading products using the Jiangbo chart to display upward or downward trend charts. Figure 10: An example of the function/interface implementation of the method for presenting future trend lines for trading products according to the present invention in the application of trading products using the Jiangbo chart to display horizontally symmetrical trend charts. Figure 11: An example of the application of the custom trend line generated by this invention to the operation/adjustment method of a type of computer.

1: Endpoint Finding Algorithm 11: Endpoint Data 2: Critical Point Algorithm 21: Critical Point 3: Trend Line Algorithm 31: Vertex-to-Vertex Connection 32: Vertex-to-Reversal Connection 33: Endpoint-to-Endpoint Connection 34: Slope Filtering 35: Spacing Filtering 4: Trend Line Drawing or Output Completed

Claims (9)

A trend line method includes the following steps: an endpoint finding algorithm that provides at least one endpoint data; the endpoint finding algorithm refers to a method for establishing high/low endpoints; it obtains endpoints by sorting the data of each point, initiating the calculation based on the highest/lowest value, distinguishing the left and right intervals based on the highest/lowest value, and continuously indexing the descending/ascending sort of each time point; A critical point algorithm is connected to the endpoint finding algorithm. The endpoint data obtained by the endpoint finding algorithm is filtered by the critical point algorithm to select at least one critical point data. The critical point data includes at least one vertex and at least one inversion point. The critical point algorithm refers to finding the correlation between the high and low values of the endpoint data and determining whether each endpoint is a vertex or an inversion point according to the time sequence time point value, so as to complete the calculation operation of defining critical points such as vertices and inversion points. A trend line algorithm is connected to the critical point algorithm. The trend line algorithm obtains the data of the endpoints and critical points from the endpoint search algorithm and the critical point algorithm, and connects the critical points to form a trend line to complete and output the trend line. A trend line drawing or output completion is completed by connecting the trend line algorithm and completing the drawing or output of the trend line. The trend line output includes, but is not limited to, a trend line data table, a trend line matrix data table, a trend line chart, etc. The trend line is generated by predicting and calculating the endpoints of the current trading chart area. New endpoints can be continuously generated or old endpoints can be eliminated based on the latest real-time data. Future trend lines can be continuously built in this way. Trend lines of different time levels can be applied to the window of the time level in effect. As described in the trend line method of claim 1, the endpoint data and the critical point data may be endpoint, vertex, and inversion point data in the Critical Point Data Table (CPT). The trend line method as described in claim 1, wherein the trend line algorithm can form a trend line that includes, but is not limited to, a vertex-to-vertex form, a vertex-to-reversal form, an end-to-endpoint form, a slope-based filtering form, and a spacing-based filtering form. As described in claim 3 regarding the trend line method, the peak-to-peak connection method involves connecting peaks in the Critical Point Table (CPT) to form a trend line that can represent both upward and downward trends; the peak-to-reversal connection method involves connecting peaks in the CPT to form a trend line that can represent both upward and downward trends; the endpoint-to-endpoint method involves connecting peaks in the CPT to form a trend line that can represent both upward and downward trends. The trend line formed by connecting the endpoints of the boundary point data table (CPT) to the vertex and inversion point; the slope filtering method is to filter the trend lines in the trend line data table (TT) according to their slope being close to a certain slope or having a large difference; the interval filtering method is to filter the trend lines in the trend line data table (TT) when their endpoint values are greater than or less than a certain interval. As described in claim 1, the trend line method, wherein the completed trend line output is output to the program system interface of the trading product, and can be displayed dynamically as the trading time progresses through a subsequent trend line prediction area, a trend line control area, and an endpoint prompt control area. As described in claim 1, the trend line method, wherein the endpoint finding algorithm, the critical point algorithm, and the trend line algorithm are established by including but not limited to node data, hash data, and a critical point data source to establish endpoints, critical points, and trend lines; wherein the node data refers to the relevant records of the main nodes, including but not limited to time point index, time point, K-bar's highest value, lowest value, closing value, opening value, candidate record value, left high point record, right high point record, left low point record, right low point record, etc.; the hash data refers to the data obtained through a hash function (Hash). The function (HF) is used to calculate the position corresponding to a key and a value; the critical point data refers to the points within the trend line endpoints, including but not limited to high points, peaks, and reversal points. As described in claim 1, the trend line method, wherein the completed trend line drawing or output data table includes, but is not limited to, a trend line data and a trend line matrix data; the trend line data refers to the data within each region divided by the critical point, the vertex, and the reversal point. The data used to generate trend lines is combined within the intervals. This trend line data includes, but is not limited to, the types of trend lines from upper peak to lower peak, lower peak to upper peak, peak to reversal point, reversal point to peak, endpoint to peak, and endpoint to endpoint. The trend line matrix data can combine or discard the trend lines within the trend line data. The trend line method as described in claim 1, wherein the trend line method can be applied to singular or plural predecessor market conditions, and provides a method for setting multiple endpoints to form a trend line through trend line setting operation/adjustment. As described in claim 8 of the patent application, the preceding market refers to the market results of the country where the market opened for trading earlier. The preceding market is the market that provides a method for setting multiple endpoints to form a trend line, which can be used to handle single or multiple preceding market data.