JPH11224295A - Brand detection device whose market value indicates the extreme value of the brand and its use - Google Patents

Abstract

(57) [Summary] [Problem] To select stocks with a ceiling or bottom price from stocks, exchange rates, and commodity transactions from all stocks and distribute them to investors. SOLUTION: Short-term interval p + 1 and 2p + 1 ·
Based on the respective movement tendency values calculated at the intervals of 4p + 1 / 8p + 1.16p + 1, the central portion of each of the movement trends is referred to.
The upper price section (F / F2 / F3 / F4 / F5 = H) and the lower price section (F / F2 / F3 / F4 / F5 = L) are set separately, and the fluctuation of the trend value curve of the stock price trajectory reaching the market price is ( F5 +
F4 + F3 + F2 + F). Before extremum at this time, the extreme value signal (T _U · X _U) is detected (F → F2
→ F3 → F4 → F5 = H), the extremum with a longer rising period, the extremum before / extreme value signal ( _TL /
X _L ) is detected (F → F2 → F3 → F4 → F5 = L) and output and distribution is made as an extreme value brand with a long falling period.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION The present invention relates to a method for detecting stocks in which the latest market transaction value of each stock is an extreme value of the stock's high or low price in stock investment, foreign exchange, commodity trading, etc. This is a method of operating the transmission media and distributing it to specific investors, other institutions, etc.

[0002]

2. Description of the Related Art The following terms used in the present invention are defined as follows. Market; Stocks, foreign exchange, and commodity markets Stock prices; Prices, exchange rates or indices established in the market Market value; The latest one of the preceding section's stock prices Current: Time corresponding to the market price Brands: Stocks, currencies, trade names, etc. traded in the market Extreme value: The value at the point where the tangent to the stock curve is horizontal. Nearest maximum / minimum value interval; range of stock price time series Efforts have been made for a long time to narrow down investment targets and their action periods in accordance with the investment principle of “getting the maximum profit at the minimum cost”. Analysts consider fundamental conditions and offer promising investment securities through various information media. However, at the time of the investment, in other words, there is almost no information on stocks that can buy at the extremes with a low market price and those that should sell at the extremes with a high market price. On the form, many proposals have been made for the corresponding technical analysis. As representative indexes, Granville's moving average, Elliott's wave theory, psychological line, volume ratio, deviation rate from the moving average, and various other methods have been adopted (Ref. 1). Conventional technical analysis is based on the assumption that, as in Granville's law, a certain pattern is determined and that pattern is led to a future pattern.
Not only is it difficult to determine to which pattern the market value belongs, or when the pattern is formed, but it is often too late when the pattern is confirmed.

[0003] Investors and companies must always read trends in market prices, foresee their changes, and select one of selling, buying, or forgotten. Investors are paying attention to the question, "Does the price continue to rise and set a high ceiling now?" Or, "Did you keep falling downward and hit the bottom of the downside?" In other words, today it is of utmost importance to narrow down the stocks that have an extreme value at the present time, but conventional technologies cannot answer them. Conventional technology has given up on pursuing these extremes, as the industry has been saying that the head and tail are ugly.

[0004]

From the conventional research, it is almost impossible to predict future stock prices by extending stock price time series results (references 2 and 3). The stock price in the market is established between the parties who sell and buy, that is, the parties who have conflicting views on the future stock price. If there is no significant change in fundamentals, the stock will recur to its median. In a situation where technical analysis does not have the ability to predict the future of stock prices, an effective investment stance is to buy stocks whose market prices have fallen and sell them at the moment market prices have risen.
However, it has not been possible to develop a method for grasping from the stocks whose market value is immediately before the extreme value to the extreme stock at this time and detecting them from all stock groups. It is the most desired to be able to provide the above-mentioned stocks to investors from over 3,000 domestic stocks and other marketable stocks. References 1 Technical Analysis of Stock Market Nikkei 2 Introduction to "Stock Investment Theory" for Securities Man Nikkei 3 Japanese Stock Price Fluctuations Toyo Keizai Shimpo (Analysis by Volatility Fluctuation Model) 4 Introduction to Time Series Model Tokyo University Press (TIME SERIES MODELS by ACHarvey)

[0005]

[Means for Solving the Problems] The stock price structure in which the market price shows an extreme value is determined by the following: a) After the trajectory of the stock price reaching the market price has been smoothed at short intervals, and From the starting point, consistently rising (falling), rising (falling)
At which the acceleration becomes zero is the extreme value of the high value (low value). B) This section is always higher (lower) than the locus value at the starting point
And this section is classified as a high value (low value) section. When the trajectory is tracked at short-term intervals, the point at which the acceleration becomes zero in this high-value (low-value) section is the short-term extreme high-value (low value) and is the target of detection. C) The magnitude of the price range at which the stock price reaches the extreme value is influenced by the rising (falling) angle and the rising (falling) period. In order to elucidate the relationship between the two factors and the stock price trajectory, the stock price movement tendency value reaching the market price is combined with a movement tendency value (N + 1) series drawn by a 2Nth interval of a short-term interval.
The rising (falling) angle of the stock price is evaluated based on the magnitude of each movement tendency value drawn by the 2Nth interval of the short-term interval. In order to measure the universal application of each issue, each numerical value is standardized and those whose absolute value exceeds the standard are targeted. During the rising (falling) period when the stock price reaches the extreme value, the + (-) period with the same sign goes up and down from the present time when the synthesized movement tendency value reaches the extreme value and during the period. At any point in time, the stock price is the extreme of the rise (fall). D) The above is the stock price structure in which the market value shows the extreme value. The market prices of all the stocks are input and only the stocks that have passed this structure filter are taken.

[0006] The trajectory of the stock price reaching the market value is as follows: a) The market value _Dt is obtained via media such as broadcasting, communication, and printed matter. B) The time series actual value of the stock price ΔD ₁ , D ₂ ,..., D _t-2 , D recorded and stored from a storage device such as a hard desk.
_t-1 ｝ is reproduced and output. At the end section of the time-series actual value c) is added market D _t from the input device. D) The time series data is ΔD ₁ , D ₂ ,... D _t-2 , D
_t-1 , D _t }.

Basically, ΔD ₁ , D ₂ ,... D _t−2 , D
_t-1, D _t} is the mutual lag _{1 (D t- μ) = b} t * (D t-1 -μ) + ε t t = t,
t-1... 3,2. Where b _t , μ; parameter ε _t ; mean 0, variance σ-squared disturbance term Each time-series numerical value is composed of a tendency value b _t and a random disturbance term ε _t added thereon, and the tendency value b _t is considered to be a smooth function with respect to time (Ref. 4).

[0008] In order to remove irregularities in numerical values due to the disturbance term, smoothing is performed by the least square method. As shown in FIG. 2, a trend value b _t and a smoothed value y _t are obtained from a regression line in the range of p + 1 time series unit of {D _t , D _t−1 ,... D _tP }. Then, the trend value b _t-1 and the smoothed value y _t-1 in the range of {D _t-1 , D _t-2 ,... D _tP-1 } are obtained. Further, t = t
Repeat until −2, t−3,..., P + 1. The interval of p + 1 is short-term, and usually p + 1 = 3 to 7 is used.

In the procedure of the flowchart of FIG. 12, {y _t , y _t−1 ,... Y _{P + 1} }; a time series smoothed numerical value and its y curve. {B _t , b _t−1 ,... B _{P + 1} };
The b curve. {D _t , D _t−1 ,... D _{P + 1} }; A four-valued ruled line and its D curve are obtained. FIG. 5 shows the relationship between the three curves. Similarly, 2ND interval 2p + 1 is adopted, and {b2 _t , b2 _t−1 ,... B2 _{2p + 1} }; b2 curve, 3RD interval 4p + 1 is adopted {B3 _t , b3
_t-1 ,... b3 _{4p + 1} }; moving tendency value b3 curve, adopting 4TH interval 8p + 1 {b4 _t , b4
_t-1 ,... b4 _{8p + 1} }; Moving tendency value b4 curve is adopted by using 5TH interval 16p + 1 {b5 _t , b5
_t-1 ... b5 _{16p + 1} };

The short-term (p + 1) interval is, for example, 5 days = p + 1, which is one week in the case of daily activities, ie, 2p + 1 = 9 days, 3RD interval, in order to exclude the influence of seasonal and cyclical fluctuation factors. After that, 4TH interval / 5TH with (previous interval-1) * 2 + 1
Determine the interval.

The extreme value of the y-curve is the position of zero of the first-order derivative curve. As one of the creation of the first-order derivative curve, the short-term movement tendency value b curve is applied. The b curve corresponds to the first derivative curve of the y curve of the lag p / 2 from the relationship shown in FIG. That is, each time point shifted backward by p / 2 from the current time point is the time point tendency value of the y curve. Similarly, the curve b2 is p, the curve b3 is 2 * p, the curve b4 is 2 * 2 * p, and the curve b5 is 2 * 2.
Each time point shifted backward by * 2 * p is the time point trend value of the y curve. In FIG. 3, the short-term movement tendency value and 2ND to 5
13 shows the correlation between the movement tendency value curves of the TH interval.
The content of the b5 curve of the macro fluctuation of the stock price smoothed y curve is b4
The curve is shown, and the details are further overlapped with the b3 and b2 curves, and the most local fluctuation is the b curve of the short-term movement tendency value.
In other words, the b curve faithfully represents the local tendency of the y curve,
The transition from the microscopic viewpoint to the macroscopic view is made as the transition to the b2, b3, b4, b5 curve is made.

FIG. 4 is a partially enlarged view of the trend value curve portion of FIG. The b5 curve is b5E, the b4 curve is b4E, the b3 curve is b3E, and the b2 curve is unknown from b2E to the current time. b4S dropped to b4 curve from b5E
The relative relationship between b4E and b4E is replaced with the relative relationship between b5E and b4E. Similarly, the relative relationship between b3S and b3E, which is reduced to a b3 curve from b4E, is the relative relationship between b4E and b3E. The relative relationship between b3E and b2E and between b2E and bE is determined in the same manner. The change in the trend value up to this point is b
5E → (b4S to b4E) → (b3S to b3E) → (b
2S to b2E) → (bS to bE). b of the b2 curve
2E is the starting point of (bS to bE) of the b curve, b3 of the b3 curve
E is a starting point of (b2S to b2E) of the b2 curve. This transition line goes back to the present time, and in the range where the sign does not change, when the stock price y curve is consistently + (-), it rises (falls) and reaches the present time.

The trajectory of each stock's stock price shows a behavior peculiar to that stock. To measure universal application to all issues,
Normalize each variation. The moving tendency value derived from the time series stock price is normalized such that the average is zero and the variance is one. The rising (falling) angle of the stock price is the trend value itself, and only the section where the standardized trend value exceeds (below) the standard value is high (low value)
One standard deviation is used as an upper division line V _U (lower division line V _L ) as a standard.

[0014] Short-term movement tendency value in the flow chart of the procedure of FIG. _{13 {b t, b t-} 1, ··· b 1} to scale the _{{B t, B t-1} , ··· B 1} B criteria Obtain the transformation curve.
FIG. 5 shows the stock price y-curve and the B-normalized curve, and the horizontal axis represents time series.
In addition, the vertical axis scale is shown as a standard deviation. _Bt-3 , _Bt-2 , B
local change of the trend values leading to _t-1, B _t, and acceleration of the y curve, expressed emphasized. And trajectory leading to B _t is the starting point of time was exceeded upper dividing line V _U (lower dividing line V _L) (below) to further increase (reduction), inverted or deceleration (acceleration), its tendency value b is zero Is defined as the short-term upside segment F = H (downside segment F = L) of the B-based curve, and the trajectory of the stock price rises (falls) consistently from the above-mentioned starting point, and the final time point of F is an extreme value. Becomes If the time at which B _t is reached the tendency value b _t = 0 from the upside segment F = H in B reference curves and _{X U, y t-2 /} p is the maximum of the y curve, than delimiters F = L point was reached the b _t = 0 X _L denotes the minimum of y _{t-2 / p} of y curve, B standardized curve beyond from going back over dividing line V _U (lower dividing line V _L) it the folded point T _U (T _L) are both detected and the immediately preceding signal y curve give an extreme value.

{B2 _t , b2 _t−1 ,... B2 _{2p + 1} };
B2 curve is standardized and {B2 _t , B2 _t-1 ,... B2
_{2p + 1} }; Convert to B2 normalized curve. The period until the trajectory leading to B2 _t increases (decreases below) above the upper division line V _U (lower division line _VL ), reverses and decelerates (accelerates) until the tendency value b2 becomes zero, Upper section F2 = H of B2 standardized curve
(Lower price section F2 = L) and sorting. FIG. 6 shows B in FIG.
It is obtained by adding the two standardization curves and the selection result F2. B
The _2t sorting F2 _t is b2 which is p backward from the current time in FIG.
This is the classification of E and bS to the current starting point.

{B3 _t , b3 _t−1 ,... B3 _{4p + 1} };
B3 curve is standardized and {B3 _t , B3 _t-1 ,... B3
_{4p + 1} }; Convert to B3 normalized curve. Similarly, {b4 _t , b4 _t−1 ,... B4 _{8p + 1} }; b4 → {B4
_t , B4 _t-1 ,..., B4 _{8p + 1} }; B4 {b5 _t , b5 _{t-1 ,,} ... b5 _{16p + 1} }; b5 → {B
_{5 t, B5 t-1,} ··· B5 16p + 1}; converted to B5 standard curves. B3 _{t-B4 t-B5} increase locus leading to _t exceeds the upper dividing line V _U (lower dividing line V _L) and (Shitamawa' decreased), inversion and deceleration (acceleration) and - its tendency value b3 b4 -
The period until b5 becomes zero is defined by the upper section F3, F4, F5 = H (lower section F3, F5) of the B3, B4, B5 standardized curve.
4.F5 = L) and sort. FIG. 7 shows B3 · B in FIG.
4.B · 5 standardized curve and each selection result F3 · F4 · F5 are added. B3 _t · B4 classification selection of _{t · B5 t F3 t · F4} t · F5 t is 2p · 4p from the present time in FIG. 4
・ B3E, b4E, and b5E are distinguished from each other by 8p and (b2S to b2E) · (b3S to b3E) · (b4
S to b4E).

At the present time, the signals T _U (T _L ) and X _U (X
_L ) is detected and the upper section F2 _t = H (the lower section F2 _t).
= L), the signal T _U (T _L ) · X _U (X _L ) arrives at least after a rise (fall) for a period of p,
The signals T _U (T _L ) and X _U (X _L ) are T1U (T1
L) .X1U (X1L). Furthermore, F3 _t
= H (lower price category F3 _t = L), T1U (T1L) · X1 after going up (falling) for at least 2 * p period
U (X1L) is T2U (T2L) · X2U (X2
L). Further, F4 _t = H (F4 _t = L)
In the case of, T2U (T2L) and X2U (X2L) are T3U after going up (falling) for at least the period of 2 * 2 * p.
(T3L) · X3U (X3L). Further, in the case of F5 _t = H (F5 _t = L), at least 2 * 2 * 2
* T3U (T3L) after the rise (fall) of the period of p
-X3U (X3L) is T4U (T4L)-X4U (X4
L).

FIG. 1 is a diagram showing the steps after the above-described process in a time series. The numerical value of each item in the time series is derived from the actual results only before the time series item, and is sequentially added. Similarly, the extreme value in the past and the signal immediately before the extreme value are detected under the condition that the time point is the last term.

[0019]

FIG. 10 is a flow chart showing the structure and functions of the present invention. The market value D _{t + 1} for each target brand obtained through communication or other transmission media is added to the last item of the time series {D _t , D _t−1 ,... D ₁ } reproduced and output from the storage device. constitute new _{{D t + 1, D t} , D t-1, ··· D 1} a. Market value D _{t +} b _{t +} 1 commensurate with _1, B _{t + 1} to calculate the extreme front signal (T _U, T _L) · extrema signal (X _U,
X _L ) are firstly sorted out for the stocks that have seen the occurrence, and according to the flowchart shown in FIG.
The secondary detection work immediately before the extreme value and the extreme value brand is performed for all the brands by market. If there is a target brand to be detected in a market or a specific field requested by the investor in advance, the print / fax or other communication device is operated and distributed.

The stock price actual time series of {D _t , D _t−1 ,... D ₁ } is output in advance from the storage device, and {b _t , b _t−1 , by the procedure of FIG. , ... b _{p + 1} }; short-term movement tendency value b and its average value, standard deviation, {y _t , y _t-1 , ... y _{p + 1} }; Output to the storage device again.

The short-term trend value b _{t +} derived at the short-term interval p + 1 from the current value D _{t + 1} from the input device and {D _t , D _t−1 ,... D _{t− p + 2} } output from the storage device. _1, B is calculated scaled value B _{t + 1,} and B _t is more than the upper dividing line V _U at B scaled value _{B t-1 <B t>} B t + 1, its locus convex folding Time signal OP シ グ ナ ル = T _U and B _t−1 > B _t <B _{t + 1}
That is lower than the discrimination line V _L, detected its locus is the return point in time signal OP $ = T _L of concave, when the detected there is nothing OP $
= "" Is output to the brand file in the storage device.

Before the present time, the signal immediately before the extreme value OP ＄ = T
_U , OP ＄ = _TL is detected, but a brand that does not see an extreme value is detected, that is, the OP of the brand file extracted from the storage device is detected.
＄ <> “”, where the short-term movement tendency value curve (b) crosses the horizontal line, SGN (b _t ) <<> SGN (b _{t + 1} ) is OP ＄ =
T extremum of upside in the case of _{U OP $ = X U, OP} $ = For T _L and outputs the extremum OP $ = X _L and the storage device of the downside.

The brands that have seen the extreme value signal T _U (T _L ) / extreme value signal X _U (X _L ) in the primary selection of the brands are tD _{t + 1} , D _t already input to the storage device. ,
_{D t-1, ··· D 1} } from Figure 11 {b _{t +} 1 the procedure subchart sub chart 13 of Figure 12 as shown in the main _{flow, b t, b t-1} , ··· b _{p + 1} }; b short-term movement tendency value {B _{t + 1} , B _t , B _t−1 ,... B _{p + 1} };
A B-normalized value is calculated and output to the storage device.

Similarly, the 2ND interval 2p + 1 is adopted and {b2 _t , b2 _t−1 ,... B2 _{2p + 1} }; b2 curve by the procedure of the subchart of FIG. 12 and the subchart of FIG.
{B2 _{t + 1} , B2 _t , B2 _t−1 ,... B2 _{2p + 1} }; B
Store the normalized value in the storage device. 3RD interval 4p + 1.4TH interval 8p + 1.5TH sequentially in the same procedure
Repeat at interval 16p + 1 {b3 _{t + 1} , b3 _t , b3 _t-1 ,... B3 _{4p + 1} }; b
3 movement tendency value {B3 _{t + 1} , B3 _t , B3 _t-1 ,... B3 _{4p + 1} }; B
3 normalized values {b4 _{t + 1} , b4 _t , b4 _t-1 ,... B4 _{8p + 1} }; b
4 moving tendency value {B4 _{t + 1} , B4 _t , B4 _t-1 ,... B4 _{8p + 1} }; B
4 standardized values {b5 _{t + 1} , b5 _t , b5 _t-1 ,... B5 _{16p + 1} };
b5 moving tendency value {B5 _{t + 1} , B5 _t , B5 _t-1 ,... B5 _{16p + 1} };
The B5 normalized value is output to the calculation storage device.

The sub-chart of FIG. 15 shows a flow of discriminating the upper (lower) section of each moving tendency value curve. ｛B2 _t , b2
_t-1 ,... b2 _{2p + 1} }; b2 curve and {B2 _t , B2
_t−1 ,... B2 _{2p + 1} }; a locus that reaches the current time by the combination of the B2 standardized curves, and the B2 standardized curve is the upper demarcation line V _U
The period from the time of increasing (decreasing below) exceeding (lower dividing line V _L ), reversing and decelerating (accelerating) until the tendency value b2 becomes zero is represented by the upper value class F2 = H (B2 standardized curve). Down price category F2
= L) and {F2 _t , F2 _t−1 ,... F2 _{2p + 1} }; F
A two-section time series is selected and output to a storage device.

The F3 section time series is selected by combining the b3 curve and the B3 standardized curve, and the b4 curve and the B3 curve are stored in the storage device.
By the combination of four standardization curves, F4 classification time series is selected,
The F5 section time series is selected in the storage device by the combination of the b5 curve and the B5 standardized curve, and is stored in the storage device.

Stock price actual time series D; smoothed stock price y; each trend value b, b2, b3, b4, b5; each standardized curve B, B2
・ B3 ・ B4 ・ B5; F section time series F ・ F2 ・ F3 ・ F
4.F5 is taken out from the storage device, and (D
+ Y + b + b2 + b3 + b4 + b5 + B + B2 + B3 +
(B4 + B5 + F + F2 + F3 + F4 + F5), and the data is sequentially output to the storage device and the output device in chronological order until the present time.

FIG. 14 is a subchart showing a process of detecting a rising (falling) period until the market value reaches Dt _{+ 1} . The magnitude of the price range at which the stock price reaches the extreme value depends on the length of the rising (falling) period after the rising (falling) angle exceeds the standard. If the last item of the process of successively outputting is (F2 = H) in the signal immediately before the extreme value (T _U ) detection brand, (OP _U = T _U ) →
(OP ＄ = T1U), more than p period until the market value is reached,
Rising stocks. In the case of (F2 · F3 = H), (OP ＄ = T1)
U) → (OP ＄ = T2U), stocks rising for 2p or more before reaching market value. In the case of (F2 · F3 · F4 = H), (O
P ＄ = T2U) → (OP ＄ = T3U), 4 until the market value is reached
Stocks rising for more than p periods. (F2 ・ F3 ・ F4 ・ F5 =
H), (OP ＄ = T3U) → (OP ＄ = T4)
U), as a brand immediately before the rising extreme value for at least 8p periods until the market value is reached, and output to the storage device and the output device.

A signal detected immediately before the extreme value (T _L )
When the last term of the sequential output process is (F2 = L), (O
Converted from P _L = T _L ) to (OP ＄ = T1L). (F2 · F3 = L),
(OP ＄ = T1L) → (OP ＄ = T2L), down for 2p period or more before reaching market value. (F2 ・ F3 ・ F4 = L)
In the case of (OP ＄ = T2L) → (OP ＄ = T3L), the stock will be falling for 4p period or more until the market value is reached. (F2 ・ F3 ・ F
In the case of 4 · F5 = L, (OP ＄ = T3L) → (OP ＄)
= T4L), and a brand immediately before the falling extreme value is output to the storage device and the output device for at least 8p periods until the market value is reached.

In the case of the extreme signal (X _U ) detection brand, if the last term of the process of successively outputting is (F2 = H), (OP ＄ =
_{X U) → (OP $ =} X1U) to conversion, up to the market value p period or more, rising stocks. In the case of (F2 · F3 = H), (OP
＄ = X1U) → (OP ＄ = X2U), 2p until market value
Over the period, rising stocks. In the case of (F2 · F3 · F4 = H), (OP ＄ = X2U) → (OP ＄ = X3U), a stock that rises for 4p or more before reaching market value. (F2 ・ F3 ・ F4 ・
In the case of F5 = H, (OP ＄ = X3U) → (OP ＄ = X
4U), until the market value is reached, the rising extreme value brand is output to the storage device and the output device for the period of 8p or more.

In the case of the extreme signal (X _L ) detection brand, if the last term of the process of successively outputting is (F2 = L), (OP ＄ =
X _L ) → (OPＸ = X1L), down for at least p periods until the market value is reached. In the case of (F2 · F3 = L), (OP
＄ = X1L) → (OP ＄ = X2L), 2p until market value
Down for more than a period. In the case of (F2 · F3 · F4 = L), (OP ＄ = X2L) → (OP ＄ = X3L), a stock that goes down for 4p or more before reaching the market price. (F2 ・ F3 ・ F4 ・
In the case of F5 = L, (OP ＄ = X3L) → (OP ＄ = X
4L) Until the market value is reached, a descending extreme brand is output for at least 8p periods to the storage device and the output device.

The above detection process is repeated for all brands,
Signals when the market value is extreme (T1U / T2U / T3U / T4
U) is detected and the stocks before the extreme value in the storage device are stored in the file before the extreme value (T1L / T2L / T3L / T4).
L) The detected brands are stored in the brand file before the lower extreme value of the storage device,
The extreme signal (X1U / X2U / X3U / X4U) detected brands are input to the upper extremal brand file of the storage device, and the extreme signal (X1L / X2L / X3L / X4L) detected brands are input to the lower extremal brand file of the storage device. I do. If there is a detected brand in each file, it is distributed by operating a communication device such as fax and Internet to investors and other specified persons.

[0033]

[Embodiment] FIG. 1 is a distribution material for detecting an extreme value of a stock.
This is an example in which 4L, that is, a downward trend 8p = a low price extreme brand of 32 days or more is detected. Four days earlier, the signal T4L immediately before the extreme value was detected, and the investor received a report and began studying. The attention period for this issue is from the signal just before the extreme to the extreme signal.

FIG. 8 shows an example of applying the dollar / yen exchange rate in the exchange market. Immediately after the detection of the signals T2L and T3L immediately before the extreme value, the extreme value signal X3L, that is, the downward extreme value of the downward trend 4p = 16 days or more is detected, and thereafter, there is a slight rebound. Market value is not detected.

FIG. 9 shows an application example of the Tokyo soybean market in the commodity market. Signal T4L just before the extreme, that is, downtrend 8p
= In the state immediately before the lower extreme value of 32 days or more, there is a high possibility that the extreme signal X4L will be detected tomorrow from the state immediately before the bE crosses the horizontal line.

[0036]

According to the conventional research, it is almost impossible to predict the future stock price by extending the stock price time series results (references 2 and 3). In a situation where technical analysis does not have the ability to predict the future of stock prices, an effective investment stance is to buy stocks whose market prices have fallen and sell them at the moment market prices have risen. According to the present invention, it is also possible to grasp and invest at the present time from the brand immediately before the extreme value to the extreme value brand.
In addition, the ability to provide the above-mentioned sorted / understood issues to investors from over 3000 domestic stocks and other market-issued issues is to keep all issues under monitoring for a limited number of issues selected for investors.

[Brief description of the drawings]

FIG. 1 is a stock detection process report with a market price of extreme value, which is also a delivery material.

FIG. 2 is an explanatory diagram showing a correlation between a smoothed value y and a short-term movement tendency value b (differential coefficient) at a stage of smoothing an original time-series numerical value D;

FIG. 3 is a diagram showing a relationship between a stock price and each movement tendency value calculated from a short-term interval to a 5TH interval.

FIG. 4 is an enlarged view of a portion related to the movement tendency value in FIG. 3;

FIG. 5 is an explanatory diagram showing a correlation between a smoothed value y curve, a short-term movement tendency value b curve, a B-based normalized curve, and a section symbol F.

FIG. 6 adds a tendency value b2 curve, a B2 standardized curve, and a classification symbol F2 to FIG. FIG.

FIG. 7 is a correlation explanatory diagram of a stock price, a B-B5 standardized curve, F-F5 classification symbols, a signal immediately before an extreme value, and an extreme value signal.

FIG. 8 is an example of a distribution document applied to an exchange market in a stock detection process report having a market price of an extreme value.

FIG. 9 is an example of a distribution document applied to a commodity market in a report on a brand detection process in which a market price is an extreme value.

FIG. 10 is a development view of the configuration and functions of the present invention.

FIG. 11 is a general flowchart of a detection process.

FIG. 12 is a flowchart for calculating a movement tendency value.

FIG. 13 is a flowchart of standardization.

FIG. 14 is a sorting flowchart for each of immediately preceding extreme value and extreme value brand periods.

FIG. 15 is a section determination flowchart for section F.

FIG. 16 is a diagram showing a description of symbols and a relationship between symbols.

Claims (4)

[Claims] 1. A) A short-term moving tendency value time series (b) and a standardized time series (B) derived at a short-term interval p + 1 are calculated from a stock price actual time series (D) extracted from a storage device and stored in a storage device. (B) a short-term interval from a time-series numerical value (D) obtained by adding the market value D _{t + 1} via the input device to the last term of the stock price performance reaching the market value of the stock taken out from the storage device; Calculate the short-term trend value b _{t + 1} derived from p + 1 and the B normalized value B _{t + 1} ,
The normalized value B _t-1 <B _t > B _{t + 1} and B _t are the upper division line (V
_U ), the trajectory of which is a convex turning point signal OP
＄ = (T _U ), and _Bt-1 > _Bt < _{Bt + 1, that} is, below the discrimination line (V _L ), the locus of which is concave, and the signal OP 折 り = (T _L ) is detected and detected. If there is no OP は = ""
And a primary sorting process, a brand immediately before the short-term extreme value to be output to the storage device, and c) a signal OP ＄ = (T _U ) immediately before the extreme value before the current time.
OP ＄ = (T _L ) is detected, but the extreme value has not yet been detected. In other words, the short-term movement tendency value curve (b) crosses the horizontal line at the time of OP ＄ <> “″ retrieved from the storage device, SGN (b _t )
<<> SGN (b _{t + 1} ) and when OP ＄ = (T _U ), the upper extreme value OP ＄ = (X _U ), and when OP ＄ = (T _L ), the lower extreme value OP ＄ = (X _L ), the short-term extremum brand to be output to the storage device, the primary sorting process, and d) the moving tendency value b2 curve (b2) derived from the stock price performance time series (D) at the interval 2p + 1, and normalized and converted. The B2 standardized curve (B2) is calculated and output to the storage device, and the interval 4p + 1 / 8p + 1.16p +
The movement tendency value b3, b4, b5 curve (b3.
b4, b5), B3, B4, B5 normalized curves (B3, B
E) calculating and outputting to the storage device; e) each normalization curve (B, B2, B3, B4, B5) increases beyond the upper parting line (V _U ) to invert / decelerate The period until each of the tendency values (b, b2, b3, b4, b5) becomes zero is determined by the respective upper value division (F, F2, B2) of the (B, B2, B3, B4, B5) standardized curve. F3, F4, F5 = H), the period below the lower division line (V _L ) decreases, reverses, accelerates, and the period until each of the tendency values (b, b2, b3, b4, b5) becomes zero, (B, B2, B3, B4, B5) F time series (F) sorting process with the lower price segment of each standardized curve (F, F2, F3, F4, F5 = L); D) ・ Smoothing stock price (y) ・ Each trend value (b ・ b2 ・ b3 ・ b4 ・ b5) ・ Each standardized curve (B
・ B2 ・ B3 ・ B4 ・ B5) ・ F-segment time series (F ・ F2
· F3 · F4 · F5) was taken out from the storage device, wherein each time point _{D t + y t + b t} + b2 t + b3 t + b4
_{t + b5 t + B t +} B2 t + B3 t + B4 t + B5 t +
Reclassified format constituted by _{F t + F2 t + F3 t} + F4 t + F5 t, sequence t =. 1 to time up to the present time
t + 1 successive Te law in a step of sequentially outputting said storage device and an output device, g) immediately before the signal (T _U) primary detection stocks extreme, when the last term in the process of sequentially outputs of (F2 = H) , (OP ＄
= T _U ) → (OP ＄ = T1U), and if the stock price rises for more than p periods until the market value is reached (F2 · F3 = H),
(OP ＄ = T1U) → (OP ＄ = T2U), up to 2p period until market value is reached, (F2 ・ F3 ・ F4 = H)
In the case of (OP ＄ = T2U) → (OP ＄ = T3U), the rising stock / (F2 · F3 · F)
4. In the case of F5 = H, (OP ＄ = T3U) → (OP ＄)
= T4U), an output step of outputting to the storage device and the output device as a brand immediately before the rising extreme value for at least 8p periods until the market value is reached, and g) a signal immediately before the extreme value (T _L ) which is a primary detection brand and sequentially output. (F2 = L), (OP ＄
= T _L ) → (OP ＄ = T1L), p until the market value is reached
For more than the period, if the descending brand (F2 / F3 = L), (O
P ＄ = T1L) → (OP ＄ = T2L), 2 until market value
In the case of (F2 ・ F3 ・ F4 = L) for more than p period, if (F2 ・ F3 ・ F4 = L), (OP） = T2L) → (OP ＄ = T3L). F4 ・
In the case of F5 = L, (OP ＄ = T3L) → (OP ＄ = T
4L) a step of outputting to the storage device and the output device a brand immediately before the falling extreme value for at least 8p period until the market value is reached, and a) a step of sequentially outputting the extreme value signal (X _U ) primary detection brand. When the last term is (F2 = H), (OP ＄ =
X _U ) → (OP ＄ = X1U), rising stock / (P2 / F3 = H) until market value
＄ = X1U) → (OP ＄ = X2U), 2p until market value
For more than the period, rising brands (F2, F3, F4 = H), (OP ＄ = X2U) → (OP ＄ = X3U), rising brands for more than 4p periods until the market value is reached (F2, F3, F4)・
In the case of F5 = H, (OP ＄ = X3U) → (OP ＄ = X
4U), until reaching the market 8p period or more, the steps of the rising extreme value stocks to output to the storage device and an output device, j) in the extreme signal (X _L) Detection stocks, the last term in the process of sequentially outputs In the case of (F2 = L), (OP ＄ = X _L ) →
(OP ＄ = X1L), more than p period until the market value is reached,
In the case of falling brands (F2 · F3 = L), (OP ＄ = X1
L) → (OP ＄ = X2L), if the stock is falling for 2p or more until the market value is reached (F2 ・ F3 ・ F4 = L), (O
P ＄ = X2L) → (OP ＄ = X3L), 4 until market value
For more than p period, descending brands (F2, F3, F4, F5 =
L), (OP ＄ = X3L) → (OP ＄ = X4
L), a step of outputting to the storage device and the output device a falling extreme value brand for at least 8p periods until the market value is reached. . 2. An input device for inputting a current price obtained through an information medium, and b) a past time-series numerical value (D) of the brand reproduced from a storage device and a signal immediately before an extreme value (OP ＄ = T _U · T _L ), and the pre-extremum / extreme stock primary selection tendency values (b _{t-1 to} b _{t + 1} ) and the standardized values (B _{t-1 to} B _{t + 1} ) from the market value ) To calculate and exclude non-applicable brands, and a storage device to calculate and sort; c) a storage device to fill in non-applicable brands with a non-applicable mark (OP # = "") and store them again; value before and extreme value signal _{(OP $ = T U · T} L · X
_U・_XL ) Movement trend time series of primary detection brand (b5 ・ b
4 ・ b3 ・ b2 ・ b) ・ Standardized value (B5 ・ B4 ・ B3)
・ B2 ・ B) ・ Same value (F5 ・ F4 ・ F3 ・ F2 ・
F) a storage device for calculating and temporarily saving, a computing device for sequentially reading out in a time series and selecting (OP ＄ =) for each ascending and descending period, and e) a detection result (OP ＄ =) and a moving tendency for outputting the process. Value time series (b5, b4, b3, b2, b), same standardized value (B5, B4, B3, B2, B), same value (F5, F)
(4) F3, F2, and F) in the brand file and a storage device for storing the brand file; f) After all brands have been detected, the collected brand group file immediately before the upper extreme value and the brand group immediately before the extreme extreme value are collected. A storage device for retrieving a file, an extreme high-end issue group file, and a low-end extreme issue group file; g) a device for creating and outputting detection results (OP ＄ =) and its progress report from the brand group file; An extreme value brand detection / distribution device comprising a delivery / transmission device operated according to the presence / absence of a detected brand by another / detection type or by a field requested by an investor. 3. The movement tendency value (b.b2.b3.b4.b) derived from the change of the stock price time series from the short-term interval p + 1 and 2p + 1.4p + 1.8p + 1.16p + 1.
5), the change of the tendency value up to the present time is expressed by (ｂb5
E) → (b4S-B4E) → (b3S-b3E) → (b
2S-b2E) → (bS-bE), the relative relation of each adjacent division zone is determined by the upper / lower division (F5, F4, F3, F2,
In F), an output method in which a time series is displayed in a time series and clearly indicated for each ascending / descending duration, the extreme value, and the time immediately before the same are specified. 4. A) a step of inputting a market value in the called stock price actual time series; and b) an extreme value immediately before / extreme value signal (OP ＄ = T _U・ T _{L ×} X)
_U · _XL ) The detected value and the non-detected value are compared with the trend value (b _t-1 ).
b _{t + 1} ) a step of primary sorting in relation to the standardized value (B _{t-1 to} B _{t + 1} ); and c) a rise in the market value of the detected stock until it reaches an extreme value and immediately before the same. -Each trend value time series (b to b) for limiting the falling period
5) Calculate the same normalized values (B to B5) and the upper and lower standardized values (F to F5), for each period (OP ＄ = X1U to X4)
U, X1L to X4L, T1U to T4U, T1L to T4
L) A step of outputting to each file in the storage device after classification, and d) If the detected brand is present in each file in the storage device after performing all brands, the name of the detected brand and the progress of detection are output. Alternatively, a machine-readable recording medium recording a program for executing the step of delivering to a specific person.