TWM584500U - Stock-bond ratio trend prediction system - Google Patents

Abstract

A stock-to-debt ratio trend prediction system is applicable to the operation strategy for predicting the next trading day of multiple historical stock-to-debt ratio data. The prediction system includes a processor, a storage module, and an output module that are signal-connected to each other. In this way, the processor can be used to establish a survival probability model, estimate the possible development trend of the stock-to-debt ratio, and provide an appropriate operating strategy.

Description

Equity-debt ratio trend prediction system

This creation is related to a financial commodity investment forecasting related technology, especially a stock-to-debt ratio trend prediction system.

With the improvement of human living standards, in addition to obtaining active income through work, more and more people have obtained passive income through investment and financial management. Eligible investment targets are endless, and various types of financial products (such as stocks, bonds, funds, etc.) are relatively popular because of their relatively abundant public information.

In addition to investing in a single financial product, some people use composite asset portfolios to improve their investment performance. The asset portfolio can include stocks with higher risks and bonds with lower risks.

In order to be able to effectively predict the ups and downs of asset portfolios including stocks and bonds, in one embodiment, a stock-to-debt ratio trend prediction system is applicable to an operation strategy for predicting the next trading day of multiple historical stock-to-bond ratio data. The prediction system includes a processor, a storage module, and an output module that are signal-connected to each other. The processor obtains the historical stock debt ratio data from the storage module, and each historical stock debt ratio data includes multiple characteristic values. The processor uses these historical stock-to-debt ratio data to establish a survival probability model. The survival probability model includes a historical probability of each historical stock-to-bond ratio data in multiple predicted trends, and the highest survival probability of each historical stock-to-bond ratio data. The corresponding forecasting trend is the selected forecasting trend of historical stock-to-debt ratio data. The processor determines the operation strategy for the next trading day based on the selected forecast trends of the historical stock debt ratio data.

In one or more embodiments, the processor further establishes a survival probability model based on the historical stock-debt ratio data and at least one market information. Furthermore, the processor uses the historical stock-debt ratio data and at least one market information as input values of the random survival forest algorithm to establish a survival probability model.

In one or more embodiments, the processor further divides the historical stock debt ratio data into multiple first historical stock debt ratio data and multiple second historical stock debt ratio data according to different conditions of each characteristic value. And the processor establishes a survival probability model based on the first historical stock-debt ratio data. Furthermore, the processor obtains a parent confusion matrix and multiple sub-sample confusion matrices based on the second historical stock-debt ratio data, where the parent confusion matrix has a parent accuracy rate corresponding to the selected prediction trend, and each sub-sample confusion matrix corresponds to the selected The predicted trend has sub-sample accuracy. The processor also judges that each of the first historical stock-debt ratio data meets the clustering conditions. The clustering conditions should correspond to the selected sub-sample confusion matrix in the sub-sample confusion matrix. The selected prediction trend of the higher accuracy rate of the parent sample and the selected sub-sample confusion matrix is the selected prediction trend of each first historical stock-debt ratio data.

In one or more embodiments, when the highest survival probability in each first historical stock-to-bill ratio data is greater than a high threshold, the processor determines that each first historical stock-to-bill ratio data meets the clustering condition. Further, the high threshold is between 0.6 and 0.9.

In one or more embodiments, when the survival probability of each of the first historical stock debt ratio data corresponding to one of the predicted trends is greater than a low threshold, the processor determines that each of the first historical stock debt ratio data meets Grouping conditions. Further, the low threshold is between 0.4 and 0.5.

In one or more embodiments, the processor further obtains the selected weighted average of the selected forecast trends corresponding to the multiple recent stock-to-debt ratio data before the next trading day of the historical stock-to-debt ratio data to obtain The strength values of the recent stock-to-bond ratio data, and the processor further determines the next trading day based on the strength values of the recent stock-to-bond ratio data and the standard deviation of the selected forecast trends of the recent stock-to-bond ratio data. Operational strategy.

Based on the above, according to the equity-to-debt ratio trend prediction system described in one or more embodiments of the present invention, a survival probability model is used to estimate a possible trend and provide an appropriate operation strategy. In some embodiments, historical stock debt ratio data can be divided into first historical stock debt ratio data and second historical stock debt ratio data, and the first historical stock debt ratio data is used as training data to enter the model, and the first The historical stock debt ratio data is used as the verification data to generate the parental confusion matrix and the sub-sample confusion matrix, and the data of the first historical stock debt ratio data is context-sliced through clustering conditions to improve the prediction accuracy. Furthermore, in one or more embodiments, the output signal can be made smoother by using a moving weighted average method. Therefore, it is not necessary to frequently change the ratio of the stock and debt, which can effectively save transaction costs and thereby improve profitability.

As shown in FIG. 1, it shows a stock-to-debt ratio trend prediction system (hereinafter referred to as prediction system 80) according to an embodiment of the present invention, which is applicable to an operation strategy for predicting the next trading day of multiple historical stock-to-debt ratio data. In other words, the prediction system 80 can provide an operation strategy for the next trading day, which can be active, stable, or conservative, and the user adjusts the allocation of the ratio of stocks and bonds held according to the provided operation strategy. The stock-to-debt ratio information includes the transaction date and the stock-to-debt ratio. The stock-to-debt ratio is the value of the stock's return divided by the bond's return. When the stock-to-bond ratio trend is rising, it means that the stock ratio should continue to be increased and the bond ratio should be reduced for active operation; on the other hand, when the stock-to-bond ratio is trending down, it means that the bond ratio should be increased and the stock ratio should be reduced for conservative operations.

As shown in FIG. 1, the prediction system 80 includes a processor 81, a storage module 82, and an output module 83. The processor 81, the storage module 82, and the output module 83 are connected to each other through a wired or wireless signal. For example, the prediction system 80 may be an industrial computer, a personal computer, a notebook computer, a smart phone, a tablet computer, or the like. Here, the processor 81 may be implemented by one or more processing elements. Here, each processing element may be a microprocessor, a microcontroller, a digital signal processor, a microcomputer, a central processing unit, a field programmable gate array, a programmable logic device, a state device, a logic circuit, an analog circuit, a digital circuit, and / Or any device that operates signals (analogs and / or digits) based on operation instructions, but it is not limited here. The storage module 82 may be implemented by one or more storage elements. Each storage element may be, for example, a non-volatile memory, a hard disk, an optical disk, or a magnetic tape, but it is not limited thereto. In addition, it should be noted that the data stored in the storage module 82 can come from a terminal connected via a wired or wireless connection, and can be updated in real time or in batches. The output module 83 may be a screen, a printer, or a voice output device (such as a speaker), but it is not limited thereto. The following is a further description of the function of each element in this prediction system.

First, the processor 81 obtains historical stock debt ratio data from the storage module 82. Stock-to-debt ratio information is used to reflect the stock-to-debt ratio holding level of a certain investment portfolio. Among them, each historical stock-debt ratio data includes multiple characteristic values. The characteristic value includes the stock holding ratio of this investment portfolio and corresponding technical indicators (such as, but not limited to, MACD, KD, Bollinger Bands, RSI, etc.). Higher stock holdings reflect lower bond holdings and a more aggressive investment strategy; lower stock holdings reflect higher bond holdings, and more conservative investment strategies; if stocks and bonds Equity ratios make investment strategies more robust.

Then, the processor 81 uses the historical stock-debt ratio data to establish a survival probability model. The survival probability model includes the historical odds of the historical stock-to-debt ratio data in uptrend, downtrend, and consolidation trends, respectively. The predicted trend corresponding to the highest probability of survival in each historical stock-to-bond ratio data is the selected forecast trend for each historical stock-to-bond ratio data. For example, Table 1 below is taken as an example. Table 1 shows a part of the historical stock-to-debt ratio data obtained through the survival probability model, including the historical stock-to-debt ratio data for each up, down, and consolidation. The survival probability of the trend, and the trend corresponding to the person with the highest survival probability is the selected predicted trend. Taking the data on November 1, 2018 as an example, the survival probability of the stock-to-debt ratio data on that day in the downward trend is 0.18666667, the survival probability in the consolidation trend is 0.34666667, and the survival probability in the upward trend is 0.46666667. Therefore, the selected forecast trend of the stock-to-debt ratio data on November 1, 2018 is upward, and the rest of the situation will not be repeated by analogy.

[Table 1]
transaction date Selected Forecast Trends Down chance Consolidation probability Up chance 20181101 1 0.18666667 0.34666667 0.46666667 20181102 0 0.21333333 0.41666667 0.37000000 20181105 1 0.28000000 0.35666667 0.36333333 20181106 1 0.21333333 0.28000000 0.50666667 20181107 1 0.20333333 0.18000000 0.61666667 20181108 1 0.20666667 0.25333333 0.54000000 20181109 0 0.22333333 0.45000000 0.32666667 20181112 0 0.31666667 0.45666667 0.22666667 20181113 0 0.35666667 0.40666667 0.23666667 20181114 -1 0.55000000 0.19333333 0.25666667

Finally, the processor 81 determines the operation strategy for the next trading day based on the selected predicted trends of the historical stock-debt ratio data, and the output module 83 outputs the operation strategy. For example, assuming that the last piece of historical stock-to-debt ratio data is November 14, 2018, and the selected forecast trend for that day is down, the processor 81 may determine based on the selected forecast trend for November 14, 2018. The operation strategy for the next trading day (for example, November 15, 2018), and then output it through the output module 83. Specifically, when the output module 83 is a display screen, text, images, videos, or a combination of the above can be used to present the operation strategy, but it should be understood that this is not limited; as mentioned above, the output module 83 can also be a print The device can print out the operation strategy, or the output module 83 can also be a speaker and present the operation strategy in the form of sound or music. On the other hand, in this example, the processor 81 may also determine the operation on November 15, 2018 based on the average of the selected forecast trend on the day of November 14, 2018 and the days before November 14, 2018. Strategy.

In one or more embodiments of the present invention, the processor 81 further establishes a survival probability model based on the historical stock-debt ratio data and at least one market information. For example, market information can be risk sentiment, such as spreads, exchange rates, VIX indexes, etc .; market information can also be a variety of overall economic indicators, such as stock price indexes, price-earnings ratios, yields, consumer price indexes, and so on. Furthermore, in one or more embodiments, the processor 81 uses the historical stock-debt ratio data and the aforementioned market information as input values of a random survival forest algorithm to establish a survival probability model. It should be noted that although the survival probability model is established here with a random survival forest algorithm, this creation is not limited to this algorithm, as long as it can predict the survival probability.

In addition, because these historical stock-to-debt ratio data have actually occurred (for example, on November 2, 2018, the selected forecast trend was consolidation, and the actual trend was upward). Therefore, the type and / or weight of the input value of the adjustment algorithm can be used to verify the algorithm, so that the generated survival probability model can better reflect the actual situation, and then provide users with more accurate predictive analysis.

In addition, in one or more embodiments of the present invention, the processor 81 further divides the historical stock-debt ratio data into multiple first historical stock-debt ratio data according to different conditions of each characteristic value. Information on the second historical stock-to-debt ratio. For example, in an embodiment, the processor 81 divides the historical stock debt ratio data into a plurality of first historical stock debts on the condition that the transaction date of the historical stock debt ratio data is earlier than March 25, 2015. Ratio data (stock-debt ratio data with a transaction date earlier than March 25, 2015) and multiple second historical stock-debt ratio data (share-debt ratio data with a transaction date later than March 25, 2015), and the processor 81 based on these first historical stock-debt ratio data to establish a survival probability model. It should be noted that although in this embodiment, the processor 81 divides the historical stock-debt ratio data into two parts according to the condition of the transaction date, but it is not limited to this; in some embodiments, the processing The device 81 may also obtain multiple sample groups by sampling historical stock-debt ratio data, and then divide the sample group into first historical stock-debt ratio data and second historical stock-debt ratio data.

Then, the processor 81 further obtains a parent confusion matrix and a plurality of sub-sample confusion matrices based on the second historical stock-debt ratio data. The parent confusion matrix has a parent accuracy rate corresponding to the selected prediction trend, and each sub-sample confusion matrix corresponds to the selected prediction. Trends have sub-sample accuracy. In this embodiment, the parent confusion matrix is the confusion matrix obtained from all the second historical stock-debt ratio data, and it includes model prediction trends, actual trends corresponding to rising, consolidating, and falling, and average sample numbers corresponding to various situations. The sub-sample confusion matrix is a confusion matrix obtained by selecting a part of all the second historical stock-debt ratio data through different clustering conditions. It also includes model prediction trends corresponding to rising, consolidating, and falling, actual trends, and corresponding situations. The average number of samples. Specifically, the foregoing grouping condition is that when the processor 81 judges that the highest probability of each predicted trend in the second historical stock-debt ratio data is greater than a high threshold (for example, between 0.6 and 0.9), the processor 81 sets this second Historical stock-debt ratio data are grouped into Q1 to Q4 sub-confusion matrices and correspond to different scenarios. Furthermore, when the highest probability is greater than 0.9, it is grouped into the Q1 sub-confusion matrix; when the highest probability is greater than 0.8, it is grouped into the Q2 sub-confusion matrix; when the highest probability is greater than 0.7, it is grouped into the Q3 sub-confusion matrix; when the highest probability is greater than At 0.6, group to Q4 sub-confusion matrix. On the other hand, when the processor 81 judges that the probability of a selected predicted trend in the data of a second historical stock-debt ratio is greater than a low threshold (for example, between 0.4 and 0.5), the processor 81 The data are grouped into Q5 to Q11 sub-confusion matrix and correspond to different situations. Furthermore, when the probability of the selected forecasting trend is greater than 0.5, it is grouped into the Q5 sub-confusion matrix; when the probability of the selected forecasting trend is consolidation, it is grouped into the Q6 sub-confusion matrix; when the selected forecasting trend is When the probability of decline is greater than 0.5, the group is grouped into the Q7 sub-confusion matrix; when the probability of the selected forecasting trend is greater than 0.4, it is grouped into the Q8 sub-confusion matrix; when the probability of the selected forecasting trend is consolidation, the group is classified into Q9 Sub-confusion matrix; when the probability of the selected prediction trend is greater than 0.4, grouping into the Q10 sub-confusion matrix, otherwise grouping into the Q11 sub-confusion matrix.

Next, the processor 81 determines which of the clustering conditions of each of the first historical stock-to-bill ratio data determines the selected sub-sample confusion matrix corresponding to the first historical stock-to-bill ratio data, and obtains each of the first Historical stock-debt ratio data and their corresponding scenarios. In other words, the processor 81 judges which scenario each first historical stock-debt ratio data should be classified to based on the aforementioned grouping conditions, which will not be repeated here.

[Table 2]
transaction date Selected Forecast Trends Down chance Consolidation probability Up chance Situation 20181101 1 0.18666667 0.34666667 0.46666667 Q10 20181102 0 0.21333333 0.41666667 0.37000000 Q9 20181105 1 0.28000000 0.35666667 0.36333333 Q11 20181106 1 0.21333333 0.28000000 0.50666667 Q7 20181107 1 0.20333333 0.18000000 0.61666667 Q4 20181108 1 0.20666667 0.25333333 0.54000000 Q7 20181109 0 0.22333333 0.45000000 0.32666667 Q9 20181112 0 0.31666667 0.45666667 0.22666667 Q9 20181113 0 0.35666667 0.40666667 0.23666667 Q9 20181114 -1 0.55000000 0.19333333 0.25666667 Q5

After determining the context of each first historical stock-debt ratio data (ie the selected subsample confusion matrix), the processor 81 then compares the accuracy of the parent sample of the parent confusion matrix with the accuracy of the subsample of the selected subsample confusion matrix. The selection forecast trend of the higher person is used as the selection forecast trend of each historical historical debt-to-equity ratio data, and is described in Tables 3 and 4 below.

Table 3 shows the matrix confusion matrix based on a total of 2,200 stock-to-debt ratio data from June 27, 2006 to March 24, 2015. As shown in Table 3, the average number of samples (ie, the probability) that the predicted trend is down (represented by -1) and the actual trend is down is 42.57%, and the average sample that the predicted trend is down and the actual trend is consolidated (represented by 0) The average number of samples is 30.69%, and the average sample number is 24.09% when the predicted trend is down and the actual trend is up (represented by 1). Tables 4 to 14 are the 11 sub-sample confusion matrices from Q1 to Q11 based on a total of 937 stock-to-debt ratio data from March 25, 2015 to November 30, 2018. As shown in Table 12, for example, the sub-sample confusion matrix Q9 has a total of 185 samples. The average number of samples for which the predicted trend is consolidation and the actual trend is consolidation is 42.26%, while the predicted trend is upward and the actual trend is upward. The average number of samples is 35.29%. Therefore, for the equity-debt ratio data on November 7, 2018, the accuracy of the parent confusion matrix is 50.14%, which is greater than the accuracy of the sub-sample confusion matrix Q9 of 42.26%. In other words, under the premise that the predicted trend is down, the actual drop is 42.57%, the actual consolidation is 30.69%, and the actual increase is 24.09% (it is a conditional probability, not a simple probability).

[table 3]
Matrix confusion matrix actual Down (-1) Consolidation (0) Up (1) prediction Down (-1) 42.57 30.69 24.09 Consolidation (0) 18.84 50.14 31.01 Up (1) 12.94 37.06 49.65

[Table 4]
Q1 n = 1 actual Consolidation (0) prediction Down (-1) 100

[table 5]
Q2 n = 23 actual Down (-1) Consolidation (0) Up (1) prediction Down (-1) 58.82 23.53 17.65 Consolidation (0) 100.00 0.00 0.00 Up (1) 0.00 60.00 40.00

[TABLE 6]
Q3 n = 55 actual Down (-1) Consolidation (0) Up (1) prediction Down (-1) 71.43 14.29 14.29 Consolidation (0) 37.50 62.50 0.00 Up (1) 5.26 52.63 42.11

[TABLE 7]
Q4 n = 153 actual Down (-1) Consolidation (0) Up (1) prediction Down (-1) 36.96 36.96 26.09 Consolidation (0) 15.91 61.36 22.73 Up (1) 12.70 25.40 61.90

[TABLE 8]
Q5 n = 71 actual Down (-1) Consolidation (0) Up (1) prediction Down (-1) 46.48 36.62 16.90

[TABLE 9]
Q6 n = 95 actual Down (-1) Consolidation (0) Up (1) prediction Consolidation (0) 13.68 57.89 28.42

[TABLE 10]
Q7 n = 83 actual Down (-1) Consolidation (0) Up (1) prediction Up (1) 3.61 33.73 62.65

[TABLE 11]
Q8 n = 95 actual Down (-1) Consolidation (0) Up (1) prediction Down (-1) 38.71 31.18 30.11 Consolidation (0) 0.00 100.00 0.00 Up (1) 100.00 0.00 0.00

[TABLE 12]
Q9 n = 195 actual Down (-1) Consolidation (0) Up (1) prediction Consolidation (0) 22.02 42.26 35.71 Up (1) 11.76 52.94 35.29

[TABLE 13]
Q10 n = 71 actual Down (-1) Consolidation (0) Up (1) prediction Up (1) 22.54 40.85 36.62

[TABLE 14]
Q11 n = 93 actual Down (-1) Consolidation (0) Up (1) prediction Down (-1) 34.21 31.58 34.21 Consolidation (0) 13.79 48.28 37.93 Up (1) 23.08 42.31 34.62

In each sub-sample confusion matrix, the prediction result can be further adjusted according to the probability distribution of the confusion matrix. For example, if the preliminary forecast result is up and the scenario is Q9, because the forecast in the Q9 scenario is predicted to rise and the actual trend is the highest probability of consolidation (52.94%), the forecast result is changed from up to consolidation. Conversely, if the preliminary forecast result is down and the scenario is Q3, the forecast trend will remain down since the actual trend is still the highest (71.43%) because the forecast is down in the Q3 scenario.

Taking the stock-to-debt ratio data on November 7, 2018 as an example, the selected forecasting trend is down and belongs to the situation Q4. The maternal confusion matrix corresponding to the downtrend has a maternal accuracy of 42.57%, and the sub-sample confusion matrix Q4 corresponds The sub-sample accuracy of the downtrend is 36.96%. Therefore, the processor selects the selected forecasting trend of the parent confusion matrix as the selected forecasting trend of the equity-debt ratio data on November 7, 2018, that is, the downward trend.

Thereby, according to the prediction system of one or more embodiments of the present invention, the prediction result of the model can be verified and modified by using the scenario analysis method, so as to better reflect the actual situation and provide users with more accurate prediction analysis.

In one or more embodiments, the processor further obtains the selected weighted average of the selected forecast trends corresponding to the multiple recent stock-to-debt ratio data before the next trading day of the historical stock-to-debt ratio data to obtain The strength of these recent stock-to-debt ratio data. For example, assuming that the next trading day is November 15, 2018, the processor can perform a moving weighted average based on the selected forecasting trend corresponding to the stock-to-debt ratio data on the 20th day before that date to obtain nearly 20-day stocks and bonds. Than the selected forecast trend of the data (ie intensity value), when the intensity value is less than 1 but greater than or equal to 0.4, the defined intensity value is located in the first interval; when the intensity value is less than 0.4 and greater than or equal to 0, the defined intensity value is located at The second interval; when the intensity value is less than 0 but greater than or equal to -0.4, the defined intensity value is located in the third interval; and when the intensity value is less than -0.4 but greater than or equal to -1, the defined intensity value is located in the fourth interval.

Then, the processor 81 further determines the standard deviations of the selected forecasted trends of the recent stock-to-debt ratio data. When the standard deviation of the selected forecast trend is greater than 0 and less than or equal to 1, it is a low volatility, which means that the selected forecast trend has little change in these recent times of stock-debt ratio data; on the other hand, when the selected forecast When the standard deviation of the trend is greater than 1, it belongs to high volatility, which means that the selected forecast trend has a large change at these times in the recent stock-debt ratio data.

Then, the processor 81 then determines the operation strategy for the next trading day based on the strength values of the recent stock-to-debt ratio data and the standard deviations of the selected predicted trends of the recent stock-to-bond ratio data. For example, when the intensity value is located in the first or second interval and has high volatility or is located in the first interval and has low volatility, the processor 81 determines that the operation strategy for the next trading day is positive, and provides stock bonds Operational configuration with a ratio of 7: 3; when the intensity value is in the second or third range and has low volatility, the processor 81 decides that the operation strategy for the next trading day is robust, and provides a stock-to-debt ratio configuration of 5 : 5 operation suggestions; and when the intensity value is in the third or fourth interval and has high volatility or the fourth interval has low volatility, the processor 81 decides that the operating strategy for the next trading day is conservative, and Provide operational recommendations with a debt-to-equity ratio of 3: 7. In this way, the output signal can be made smoother by moving weighted average, which means that investors do not need to frequently change the ratio of stock and debt, which can effectively save transaction costs and improve profitability.

Referring to FIG. 2, FIG. 2 is a flowchart of steps of a method for predicting a stock-to-debt ratio trend according to a second embodiment of the present invention, which is applicable to an operation strategy for predicting the next trading day of multiple historical stock-to-debt ratio data. As shown in FIG. 2, one or more embodiments of the present invention further disclose a method for predicting the price trend of financial commodities, which includes the following steps: a historical share debt ratio data acquisition step S101, a survival probability model establishment step S102, and an operation strategy decision step S103 Each step is explained below.

The step of obtaining historical stock-to-bill ratio data includes obtaining the historical stock-to-bill ratio data, as shown in step S101; wherein each historical stock-to-bill ratio data includes multiple characteristic values. The description of the eigenvalues has been described before, and will not be repeated.

The step of establishing the survival probability model includes establishing the survival probability model based on the historical stock-debt ratio data, as shown in step S102; wherein the survival probability model includes the survival probability of each historical stock-debt ratio data in multiple predicted trends, and in each history The forecast trend corresponding to the highest probability of survival in the stock-to-debt ratio data is the selected forecast trend for each historical stock-to-debt ratio data.

The operation strategy determination step includes determining the operation strategy for the next trading day based on the selected predicted trends of the historical stock-to-debt ratio data, as shown in step S103.

Please refer to FIG. 3, which is a detailed flowchart of a method for predicting a stock-to-debt ratio trend according to a third embodiment of the present invention. As shown in FIG. 3, according to one or more embodiments of the present invention, in step S102 for establishing a survival probability model, the method further includes: Step S1021: establishing a survival probability model based on the historical stock-debt ratio data and at least one market information.

Please refer to FIG. 4, which is a detailed flowchart of a method for predicting a stock-to-debt ratio trend according to a fourth embodiment of the present invention. As shown in FIG. 4, according to one or more embodiments of the present invention, in step S102 of establishing a survival probability model, step S1022 is further included: the historical stock-debt ratio data and at least one market information are used as inputs of a random forest algorithm Value to build a survival probability model.

Please refer to FIG. 5, which is a detailed flowchart of a method for predicting a stock-to-debt ratio trend according to a fifth embodiment of the present invention. As shown in FIG. 5, according to one or more embodiments of the present invention, in step S102 of establishing a survival probability model, step S1023 is further included: according to different conditions of each characteristic value, the historical stock debt ratio data is divided into multiple The first historical stock debt ratio data and the plurality of second historical stock debt ratio data; and step S1024: establishing a survival probability model based on the first historical stock debt ratio data.

Further, please refer to FIG. 6, which is a detailed flowchart of a method for predicting a stock-to-debt ratio trend according to a sixth embodiment of the present invention. As shown in FIG. 6, according to one or more embodiments of the present invention, in step S102 of establishing a survival probability model, in addition to the foregoing steps S1023 and S1024, the method further includes step S1025: obtaining the parent body according to the second historical stock-debt ratio data. Confusion matrix and multiple sub-sample confusion matrices, in which the parent confusion matrix has a maternal accuracy rate corresponding to the selected prediction trend, and each sub-sample confusion matrix has a sub-sample accuracy rate corresponding to the selected prediction trend; step S1026: judging each first historical stock-debt ratio The data satisfy the clustering condition, where the clustering condition corresponds to the selected sub-sample confusion matrix in the sub-sample confusion matrix; and step S1027: according to the accuracy of the parent sample of the parent confusion matrix and the accuracy of the sub-sample of the selected sub-sample confusion matrix. The selected forecast trend of the higher one is used as the forecast trend of each first historical stock-debt ratio data.

Please refer to FIG. 7, which is a detailed flowchart of a method for predicting a stock-to-debt ratio trend according to a seventh embodiment of the present invention. As shown in FIG. 7, according to one or more embodiments of the present invention, in step S1026, step S1028 is further included: when the highest survival probability in each first historical stock-debt ratio data is greater than a high threshold value, each first judgment is made. The historical stock-debt ratio data meets the grouping conditions. Further, in one or more embodiments, the high threshold is between 0.6 and 0.9.

Please refer to FIG. 8, which is a detailed flowchart of a method for predicting a stock-to-debt ratio trend according to an eighth embodiment of the present invention. As shown in FIG. 8, according to one or more embodiments of the present invention, in step S1026, step S1029 is further included: when each first historical stock-to-debt ratio data corresponds to one of the predicted trends, the survival probability is greater than When the threshold is low, it is judged that the data of each first historical stock debt ratio meets the grouping conditions. Further, in one or more embodiments, the low threshold is between 0.4 and 0.5.

Please refer to FIG. 9, which is a detailed flowchart of a method for predicting a stock-to-debt ratio trend according to a ninth embodiment of the present invention. As shown in FIG. 9, according to one or more embodiments of the present invention, in the operation strategy decision step S103, step S1031 is further included: a plurality of recent stock-to-debt ratios before the next trading day of the historical stock-to-debt ratio data. The selected forecasting trends corresponding to the data are subjected to moving weighted average calculation to obtain the strength values of the recent stock-to-bond ratio data; and step S1032: based on the strength values of the recent stock-to-bond ratio data and the recent stock-to-bond ratios The standard deviation of the selected forecast trends of the data determines the operating strategy for the next trading day.

Based on the above, according to the equity-to-debt ratio trend prediction system described in one or more embodiments of the present invention, a survival probability model is used to estimate a possible trend and provide an appropriate operation strategy. In some embodiments, historical stock debt ratio data can be divided into first historical stock debt ratio data and second historical stock debt ratio data, and the first historical stock debt ratio data is used as training data to enter the model, and the first The historical stock debt ratio data is used as the verification data to generate the parental confusion matrix and the sub-sample confusion matrix, and the data of the first historical stock debt ratio data is context-sliced through clustering conditions to improve the prediction accuracy. Furthermore, in one or more embodiments, the output signal can be made smoother by using a moving weighted average method. Therefore, it is not necessary to frequently change the ratio of the stock and debt, which can effectively save transaction costs and thereby improve profitability.

80‧‧‧ Forecast System
81‧‧‧ processor
82‧‧‧Storage Module
83‧‧‧Output Module
S101 ~ S103, S1021 ~ S1029, S1031 ~ S1032‧‧‧Steps

[Fig. 1] Schematic diagram of the stock-to-debt ratio trend prediction system of the first embodiment of this creation.
[Fig. 2] A flowchart of steps in a method for predicting a stock-to-debt ratio trend according to a second embodiment of the present invention.
[FIG. 3] A detailed flowchart of a method for predicting a stock-to-debt ratio trend according to a third embodiment of the present invention.
4 is a detailed flowchart of a method for predicting a stock-to-debt ratio trend according to a fourth embodiment of the present invention.
5 is a detailed flowchart of a method for predicting a stock-to-debt ratio trend according to a fifth embodiment of the present invention.
6 is a detailed flowchart of a method for predicting a stock-to-debt ratio trend according to a sixth embodiment of the present invention.
FIG. 7 is a detailed flowchart of a method for predicting a stock-to-debt ratio trend according to a seventh embodiment of the present invention.
[FIG. 8] A detailed flowchart of a method for predicting a stock-to-debt ratio trend according to an eighth embodiment of the present invention.
[FIG. 9] A detailed flowchart of a method for predicting a stock-to-debt ratio trend according to a ninth embodiment of the present invention.

Claims (10)

A stock-to-debt ratio trend prediction system suitable for an operation strategy for predicting the next trading day of multiple historical stock-to-debt ratio data. The prediction system includes:
A processor, a storage module, and an output module. The processor, the storage module, and the output module are signally connected to each other. Among them:
The processor obtains the historical stock-to-bill ratio data from the storage module, wherein each of the historical stock-to-bill ratio data includes multiple characteristic values;
The processor establishes a survival probability model based on the historical stock debt ratio data, wherein the survival probability model includes a survival probability of each historical stock debt ratio data in a plurality of predicted trends, and is included in each historical stock debt ratio data The forecast trend corresponding to the highest probability of survival is a selected forecast trend for each historical stock-debt ratio data;
The processor determines the operation strategy for the next trading day based on the selected predicted trends of the historical stock debt ratio data, and the output module outputs the operation strategy. The prediction system according to claim 1, wherein the processor establishes the survival probability model based on the historical stock-debt ratio data and at least one market information. The prediction system according to claim 2, wherein the processor establishes the survival probability model by using the historical stock-debt ratio data and the at least one market information as input values of a random survival forest algorithm. The prediction system according to claim 1, wherein the processor further divides the historical stock debt ratio data into a plurality of first historical stock debt ratio data and a plurality of second history according to different conditions of the characteristic values. Stock-to-debt ratio data; the processor builds the survival probability model based on the first historical stock-to-debt ratio data. The prediction system according to claim 4, wherein the processor further obtains a parent confusion matrix and a plurality of sub-sample confusion matrices based on the second historical stock-debt ratio data, and the parent confusion matrix has a parent corresponding to the selected prediction trend. Accuracy rate, each sub-sample confusion matrix should have a sub-sample accuracy rate for the selected predicted trend; the processor judges that each of the first historical stock-debt ratio data meets a clustering condition, and the clustering condition should correspond to the sub-sample confusion matrix One of the selected sub-sample confusion matrices; the processor further makes the selected prediction trend based on a higher accuracy rate of the parent sample of the parent confusion matrix and a higher accuracy rate of the sub-sample of the selected sub-sample confusion matrix The selected predicted trend for each of the first historical stock-to-debt ratio data. The prediction system according to claim 5, wherein when the highest probability of survival of each of the first historical stock debt ratio data is greater than a high threshold value, the processor determines that each of the first historical stock debt ratio data satisfies the cluster condition. The prediction system of claim 6, wherein the high threshold is between 0.6 and 0.9. The prediction system according to claim 5, wherein when the survival probability of each of the first historical stock-debt ratio data corresponding to one of the predicted trends is greater than a low threshold, the processor judges each of the first A historical stock-debt ratio data meets the criteria for this cluster. The prediction system of claim 8, wherein the low threshold is between 0.4 and 0.5. The prediction system according to claim 1, wherein the processor further weights the selected forecasting trends corresponding to the multiple recent stock-to-debt ratio data of the historical stock-to-debt ratio data before the next trading day. An average value is used to obtain an intensity value of the recent stock-to-bond ratio data; the processor is further based on the intensity value of the recent stock-to-bond ratio data and a selected standard for the selected forecast trends of the recent stock-to-bond ratio data. The difference determines the operation strategy for the next trading day.