JPH07152725A - Function analysis method - Google Patents

Abstract

PURPOSE: To systematically simulate a tendency in the future by systematically repeating a display process corresponding to a Fibonacci ratio and the power multiple for a curve for relating two parameters with each other. CONSTITUTION: By using the maximum deviation of a parameter from the movement average of the curve 10 or an optional dynamic movement average, a first surrounding band provided with an upper band and a lower band is plotted over the entire range of the parameter to be analyzed. The upper band 32 of the first surrounding band is obtained as the sum of the maximum deviation and the lowest value indicated by one of the dynamic movement averages 22, 24, 26 and 28 or the movement average 14 along the respective points of a horizontal axis. Similarly, the lower band 34 is obtained by subtracting the maximum deviation from the highest value. Further, different two inner side surrounding bands are plotted and the inner side surrounding bands are respectively provided with the upper band 36 and the lower band 38. The two inner side surrounding bands are plotted by repeating a process for obtaining the surrounding band by using a value for which the Fibonacci ratio is repeatedly multiplied to the maximum deviation.

Description

Detailed Description of the Invention

The present invention relates to a method for systematically simulating and displaying periodic elements of the trend of a curve in order to predict future trends of the curve relating two parameters to one another.

[0002] As one of the special ways of using such a method, a financial company and other financial parties use the method to predict future trends of stocks, bonds, commodities and currencies. It is possible to give their customers an opportunity to buy or sell a certain stock or the like or a stock of a particular brand.

Another application of the method according to the invention is to anticipate future trends of particular parameters involved in various types of manufacturing processes, which parameters are above a certain critical value or below a certain critical value. Controlling the manufacturing process may be mentioned when it is important to take prompt measures to prevent

In general terms, the method according to the invention is based on the following functional analysis system. That is, in the function analysis system, a moving average (dynamic average) and a dynamic value (dynamic) similar to the conventional exponential moving average are used by using a moving average (moving average) and a fluctuation value (oscillator) according to the conventional concept.
ic moving average) and the dynamic moving average is calculated by using a Fibonacci ratio that lowers the basic moving average.

According to the invention, in a method of systematically displaying periodic elements of the movement of a curve as a schmiray in order to predict future movements of the curve relating two parameters to one another: a) Determining a moving average of the curve over one particular period of the parameter, b) calculating a difference between the actual value of the curve and the moving average, and c) Calculating an average of the variation values over the same period as for one of the parameters; d) adding the average of the variation values to the moving average to obtain a first dynamic moving average; e) determining a series of dynamic moving averages by repeating a calculation for averaging the fluctuation values over various periods in which the specific period is sequentially shortened at a predetermined ratio, and f) the moving average or Determining a maximum deviation value between all of the dynamic moving averages related thereto and calculated in advance and the curve; and g) determining a wrapping band using the maximum deviation value. The upper band of the surrounding band is the sum of the minimum value and the maximum deviation value shown in the moving average or the dynamic moving averages related thereto, and the lower band of the surrounding band is the moving average. Or the maximum value shown in its associated dynamic moving average minus said maximum deviation value, and h) further being related to said maximum deviation value by said predetermined ratio and its multiples. The above step g)
Determining the various inner and / or outer wraps by repeating the process in step 1, i.e., the process for finding the wraps, and i) the movements related to the particular time period at the predetermined ratio or multiples thereof. Sequentially repeating the process of all of the above steps with different periods of averaging, j) for each set of moving averages and their associated dynamic moving averages and said siege bands. Displaying a line; k) systematically repeating the display process in step j) according to said predetermined ratio and its multiples; l) by said step to make predictions about future trends of said curve. Evaluating the obtained information and taking action or control on matters relating to the parameter based on the evaluated information. A functional analysis method is provided.

Preferably, as the above-mentioned predetermined ratio,
The well-known Fibonacci ratio (1.618) is used, and as its multiple, the power multiple of the Fibonacci ratio is used.

Preferably, in the function analysis method according to the present invention, in order to perform the calculation and display for the various sets of lines described above, and to systematically repeat the display process described above according to the Fibonacci ratio and its powers. , Computer systems
system) is used.

When the function analysis method according to the present invention is used to predict the trend of stock prices of stocks and the like day by day, the function analysis method is used for various past periods. It is also possible to investigate the past trend of curves such as stock prices, and the prices of products closely related to them.

Next, an embodiment of the present invention will be described with reference to the accompanying drawings.

FIG. 1 is a graph for explaining how a variation value and a series of dynamic moving averages are obtained in the function analysis method according to the present invention, and FIG.
FIG. 5 is a graph showing the value of the US dollar against Deutsche Marks between 1975 and 1983, showing together five dynamic moving averages and three sieving zones obtained according to the invention, 3 and 4 show the second
FIG. 5 is a graph showing how the market actually changed after the forecast was made with the curve shown in FIG.
The figure is a block diagram showing a cooling system applicable to an industrial process and controlled by a computer.

FIG. 1 is a graph for explaining how a variation value and a series of dynamic moving averages are obtained in the function analysis method according to the present invention, and FIG.
FIG. 5 is a graph showing the value of the US dollar against Deutsche Marks between 1975 and 1983, showing together five dynamic moving averages and three sieving zones obtained according to the invention, 3 and 4 show the third
FIG. 5 is a graph showing how the market actually changed after the forecast was made with the curve shown in FIG.
The figure is a block diagram showing a cooling system applicable to an industrial process and controlled by a computer.

Referring first to FIG. 1, there is shown a curve with the reference numeral 10, which curve is plotted in magnitude on the vertical axis with respect to the time base axis. Point 12 shows the average of up to 233 plots along the horizontal axis. From such information, it is possible to determine a moving average for curve 10, which is shown by smooth curve 14. For each arbitrary point on such a moving average curve, only up to 233 plots on the curve 10 before any corresponding point on the curve 14 are calculated. The increase in the value of the curve 14 is lower than the large increase in the value of the curve 10 after the first calculation point 12. The difference 16 between the values at any two points at the same time point (the same value on the horizontal axis) is the fluctuation value (oscillator). A first dynamic moving average (dynamic move average) is obtained by adding the moving average of the variation values to the above moving average, using the average of the variation values (not shown) over a period containing the same number of plots.
ng average) 20 is obtained. Also, the number of plots related to the number of plots described above, that is, the number of plots 2
The number of plots obtained by dividing 33 by the Fibonacci factor (1.618) and its power (ie, the number of plots such as 144, 89, 55, etc.)
Using a moving average of the fluctuation values over a shorter period with
Yet another series of dynamic moving averages 22, 24 and 26 are obtained as in the previous case.

A maximum envelopment value 30 of the parameter from the moving average or any dynamic moving average is used to provide a first envelope with an upper band and a lower band over the entire range of parameters to be analyzed. -Pe) is drawn.

Referring to FIG. 2, there is shown, as a particular example, how the US dollar value fluctuated with respect to Deutsche Mark between 1975 and 1983. The calculations as described above have been carried out and in FIG. 2 a moving average 14 and a series of five dynamic moving averages 20, 2 are shown.
Curves of 2, 24, 26 and 28 and an actual value of 10 are shown.

The upper and lower zones of the first enclosure are shown as curves 32 and 34, respectively. The upper zone 32 is obtained as the sum of the minimum and maximum deviation values indicated by either the moving average 14 or the dynamic moving averages 22, 24, 26 and 28 along each point on the horizontal axis. The lower zone 34 is obtained as the maximum value shown by either the moving average 14 or the dynamic moving averages 20, 22, 24, 26 and 28, minus the maximum deviation value, along each point on the horizontal axis. . It should be noted that the various curves shown in the graph of FIG. 2 are merely drawn for the purpose of explaining the present invention so as to approximate their tendencies, and therefore the magnitudes shown by the respective curves are shown. Not accurate about that. This also applies to the graphs shown in other drawings.

Furthermore, another two inner wrapping bands are drawn,
The inner wraps have upper zones 35 and 36 and lower zones 37 and 38, respectively. Such two inner wrapping bands are used for determining the wrapping band using a value related to the maximum deviation value, that is, a value obtained by repeatedly multiplying the maximum deviation value by the Fibonacci ratio (that is, 1.618 or 0.618). It is drawn by repeating the process.

In the above example, although two inner wrapping bands are drawn, two or more inner wrapping bands can be drawn if necessary, and a plurality of outer wrapping bands can be drawn. You can also

All of the above process steps are then sequentially repeated, using different time periods for the moving average. In this case, each such period (number of plots or length of time) is related to the first period mentioned above by the Fibonacci ratio and its powers.

Having obtained a series of sieving bands over various periods of the moving average, the user can
The dynamic moving average, a series of upper and lower wraps for the initial period for the moving average, will be displayed on the video display unit of the computer system. It should be noted that such a display operation is performed by pressing an appropriate key on the key board associated with the computer. After the user has considered such display content, the user can repeat the above process by systematically pressing the key, which causes different periods of the moving average to be displayed on the video display unit. Become. In this case, the various time periods are related to the first displayed time period by the Fibonacci factor and its powers.

By carefully reviewing all the displayed information, the user can anticipate future trends in the parameters. Experience has shown that over the predictable time period over which the displayed moving averages, dynamic moving averages and bandages can be applied, such predictions can be made with very high accuracy.

As another aspect, instead of the user performing an evaluation using all the information obtained and displayed in the above-mentioned steps, the computer may be made to perform a prediction calculation about the trend of the parameter. in this case,
The user can press the appropriate keys on the keyboard to cause the computer to make the necessary forecast calculations of future trends of parameters over any desired time period, and to track such parameter trends in other ways. It can also be displayed on the video display unit together with curves such as moving average, dynamic moving average and wrapping band. As will be explained later, the computer can be programmed to evaluate the situation and make the appropriate decisions. In this case, such a decision is visibly displayed on the screen of the video display unit in such a way that the future trend of the parameter is upward, downward or stable. The computer can also be programmed to approximate the magnitude changes over time of a desired point in terms of whether the future trend of the parameter is up or down.

As will be apparent from the following description, when applying the present invention to the financial field, the user, in the light of his own experience or with the help of a computer,
It is possible to predict the movement of any one of stocks, bonds, commodities, currencies, etc. or the movement of a particular stock of them with extremely high accuracy. In this way, the user of such a system can advise his customer in deciding whether he should sell or buy the above-mentioned holdings.

The application of the present invention to the financial field will be described in more detail with reference to the examples shown in FIGS.

First, referring to FIG. 3, there is shown a graph in which prices of holdings and the like are plotted with respect to a time base axis. In this graph, a curve 10 (parameter of FIG. 1), 14 moving averages, 6 dynamic moving averages 2
0, 22, 24, 26 and 28, and two pairs of upper and lower wraps 32, 34 and 36, 38 are represented. Looking at the change in the moving average 14, it can be seen that the change over the long term changes from downward to upward. The price shown by curve 10 has picked up to the outer upper wrap 32 and all of the dynamic averages have spread well at the time indicated by line 50.

It can also be seen that the price shown with the curve 10 jumping up to the outer upper sill 32 passes through the two dynamic moving averages 29 and 28 obtained for the short period. If the holdings are bought at the beginning of the graph, that is, when the price shown by the curve 10 is less than the moving average of 14, it is possible to sell the holdings at that time and obtain a proper profit in a short period of time. it can.

FIG. 4 shows how the various curves actually changed after line 50. From FIG. 4, it can be seen that although the price has passed through the dynamic moving average 29 twice, the price has dropped sharply thereafter and is approaching the moving average 14 rapidly. In other words, it turns out that it is an opportunity for selling.

With sufficient experience, when the price of holdings etc. seems to rise or fall, it is possible to make a quick forecast by obtaining the moving average for various periods using the Fibonacci ratio and its powers. It is easy to make a purchase, and it is also easy to make an evaluation as to whether to sell or buy. All the necessary information about the various curves can be stored in the computer's memory, so that the computer itself can make a proper assessment of the situation and the expected changes in holding prices, i.e. , Or downward, and how up and down it may be shown through a visual display. Thus, the user can quickly make a decision as to whether to buy or sell a particular thing, such as a holding, with the help of the printout function of the computer.

When the present invention is applied to the industrial field, the value of a specific parameter, for example, the temperature of the liquid coolant may be continuously monitored throughout the manufacturing process or the continuous manufacturing process. In this case, the user can predict in advance when the temperature of the liquid coolant is going to exceed a predetermined critical temperature, using the process as described above, and lead to such a situation. Appropriate measures can be taken to prevent this. It should be noted that, when such a situation occurs, it is assumed that a dangerous situation or an explosive state that may cause damage and / or damage to personnel may occur.

Therefore, there are many opportunities to use the function analysis method according to the present invention in all fields of industry and also in many other fields such as the medical field in which the blood pressure of a patient is continuously monitored. is there.

Next, with reference to FIG. 5, a case where the function analysis method according to the present invention is generally applied to the industrial world will be described in detail.

Referring to FIG. 5, there is a basic layout of a system in which industrial processes are capable of systematically simulating and displaying periodic elements of liquid coolant temperature trends. Is shown as a block diagram.

In such an industrial application, the temperature of the liquid coolant is, of course, monitored so that it does not rise above or above a certain temperature, so that a defect-free product is produced. Will be gained from the process. Liquid coolant is pumped from reservoir 100 to chamber 102 by pump 104. The chamber 102 contains products, and the pump feed rate of the pump 104 is variable. In the computer 106, the temperature value signal of the liquid coolant flowing out from the chamber 102 is sent from the sensor 108 to the A / D converter 1.
Sent via 10. In the A / D converter 110, the analog signal of the temperature value is converted into a digital signal. A control unit 112 is bidirectionally connected to the computer 106. Information is provided to the control unit 112 by the manual operation unit 114 having a key board, which controls the computer 106.

The system is also provided with a video display unit (V.D.U.) 116, which video display unit has liquid coolant temperature, moving average, dynamic moving average,
As well as the various pairs of enclosures associated with the temperature of the liquid coolant. In addition, an alarm unit 118 may be provided, which makes it visible and visible in the event that the expected value for the temperature of the liquid coolant exceeds the upper limit or falls below the lower limit in the near future. And / or an audible warning.

The liquid coolant flow rate is controlled by the pump 104, and the prime mover speed of the pump 104 is controlled by the computer 106 via the D / A converter 120.

In front of the person in charge of operating the system, a video display unit 116 is placed together with a manual operation unit 114 having a key board. Thus, the operator can continuously monitor the periodic component of the liquid coolant temperature trend and also display it on the video display unit. If the person in charge of operation performs a manual operation, he / she inputs information to the control unit 112 and the computer 106 causes the D / A converter 1 to operate.
The pump 104 can be controlled via 20. Thus, if the prediction of future trends in the temperature of the liquid coolant is such that it will reach one of the above temperature limits, then the flow rate of the liquid coolant will be increased or decreased. .

On the other hand, if the operator decides to perform such control on the computer, the manual control / automatic control button on the key board of the manual operation unit 114 is operated. At this time, the above control is left to the computer 106. In this case, computer 1
At 06, the appropriate information stored in the computer's memory for the determination of moving averages, dynamic moving averages, and paired wrapping zones (which are determined for various time periods derived from the Fibonacci ratio and its powers) is used. The flow control of the liquid coolant by the pump 104 is performed according to the expectations made by the computer.

[Brief description of drawings]

FIG. 1 is a diagram showing a function analysis method according to the present invention,
3 is a graph illustrating how a variation value and a series of dynamic moving averages are obtained.

FIG. 2 is a graph showing the value of the US dollar against Deutsche Mark between 1975 and 1983, with five dynamic moving averages and three obtained according to the present invention.
3 is a graph showing two siege bands together.

FIG. 3 is a graph showing how the market actually changed after the forecast was made with the curve shown in FIG.

FIG. 4 is a graph showing how the market actually changed after the forecast shown in the curve shown in FIG.

FIG. 5 is a diagram showing a cooling system applicable to an industrial process, the cooling system being controlled by a computer system.

[Explanation of symbols]

 10 curve 14 moving average 16,30 fluctuation value 20 first dynamic moving average 22,24,26,28,29 dynamic moving average 32,35,36 upper band of the surrounding zone 34,37,38 lower side of the surrounding zone Zone 100 Reservoir 102 Chamber 104 Pump 106 Computer 108 Sensor 110 A / D Converter 112 Control Unit 114 Manual Operation Unit 116 Video Display Unit 118 Alarm Unit 120 D / A Converter

Claims (6)

[Claims] 1. A method of systematically simulating and displaying periodic elements of a trend of a curve in order to predict future trends of a curve relating two parameters to each other, comprising: a) Determining a moving average of the curve over one particular interval; b) calculating a difference between the actual value of the curve and the moving average; and c) one of the parameters. Calculating an average of the fluctuation values over the same interval as the case; d) adding the average of the fluctuation values to the moving average to obtain a first dynamic moving average; and e) selecting the specific interval. Repeating the calculation for averaging the fluctuation values over a plurality of intervals shortened at a predetermined ratio,
Determining a series of dynamic moving averages, f) determining a maximum deviation value of the curve from the moving averages and all corresponding pre-calculated dynamic moving averages, g) the maximum deviations Determining the wrapping band using a value, wherein the upper band of the wrapping band is the sum of the maximum deviation value and the minimum value represented by the moving average or the corresponding dynamic moving average. , The lower zone of the enclosure is
It is a value obtained by subtracting the maximum deviation value from the maximum value represented by the moving average or the corresponding dynamic moving average, and h) a value related to the maximum deviation value with the predetermined ratio and a multiple thereof. By repeating the formation of the bandage in step g) above to obtain a plurality of inner bandages and / or
Or an outer wrapping zone, and i) repeating all the above steps using a plurality of sections of the moving average, each section being related to the particular section with a predetermined ratio or multiples thereof. J) displaying each set of lines including a moving average, a corresponding dynamic moving average, and a wrapping band; k) the step j) according to the predetermined ratio and its multiples.
Systematically repeating the steps of: 1) evaluating the information obtained by said step to make a prediction about the future movement of said curve and, based on the evaluated information, behavior or control of matters relating to said parameter And a step of forming, and a method of function analysis characterized by including. 2. The predetermined ratio is the Fibonacci ratio (1.
618), the multiples of which are powers of multiples. 3. A computer system is used to perform calculations and displays for the plurality of sets of lines and to systematically repeat the display steps according to the Fibonacci ratio and its curtain multiples. The function analysis method according to claim 2. 4. A computer system adapted to perform all of the steps of the method for making an assessment of future developments of the curve and for printing the assessment in a suitable format. The function analysis method according to claim 3, wherein 5. The curve represents a controlled parameter in an industrial process, the parameter being kept above or below a predetermined value in the performance of the industrial process to predict future trends of the curve. 5. The method according to claim 1, wherein the parameter is output by outputting a signal when there is a possibility that the parameter is equal to or greater than or equal to the predetermined value. Function analysis method described in. 6. The signal indicating that the parameter is about to deviate from the predetermined value is an appropriate adjustment of the industrial process so that the parameter is within the predetermined value range. The function analysis method according to claim 5, wherein