JP5086952B2 - Predictive voting processor - Google Patents

Description

  The present invention relates to a technique for realizing profit distribution based on forecasts of economic indicator trends.

IT technology has promoted economic globalization, which has led to various economic indicators having a great impact not only on corporate activities but also on personal life.
JP 2002-85851 A

  In recent years, many financial products that can be accessed with a small amount of money have been developed in conjunction with various economic indicators. The profits and losses of these financial products are often determined based on whether economic indicators such as exchange rates and stock prices go up or down.

  The main object of the present invention is to provide a service in which the accuracy of the forecast for the value of the economic indicator itself determines the profit and loss.

One embodiment of the present invention relates to a predictive voting processor.
This apparatus collects votes including a predicted value of an economic index at a predetermined time from a plurality of client terminals. The “economic index” herein may be an economic index that changes as needed during the voting period, such as the Nikkei average or exchange rate. When the voting is completed and the value of the economic index at the predetermined time is determined, each vote is sorted into an out-vote that is largely out of prediction and a safe vote that is close to the prediction. Dividends will be distributed to voters of Safe Votes using the collected money as a source. At this time, the safe vote cast earlier is set so that the dividend distribution rate is higher than the later safe vote.
Note that the price, dividend, or the like mentioned here may be an actual currency, but may be a virtual currency or a currency equivalent. For example, it may be a shopping point or a lucky ticket. In any case, any value that can be measured quantitatively is acceptable.

  It should be noted that any combination of the above-described components, and the present invention expressed by a method, system, recording medium, and computer program are also effective as an aspect of the present invention.

  According to the present invention, it is possible to provide a revenue opportunity based on the accuracy of forecasts targeting economic indicators.

  In the present embodiment, it is assumed that the economic index to be predicted is “Nikkei Stock Average”.

FIG. 1 is a schematic diagram showing a transaction form in the present embodiment.
The plurality of client terminals 200a, 200b, 200c, 200d,... 200e (hereinafter simply referred to as “client terminal 200”), the bank system 210, the stock exchange system 220, and the prediction voting processing apparatus 100 communicate with each other via the Internet. They are connected to each other via a network. The client terminal 200 may be a terminal having a web browsing function, such as a personal computer or a mobile phone. Each user participates in the vote through the client terminal 200. The bank system 210 is an existing business system that realizes remittance between accounts of each user. The stock exchange system 220 notifies the Nikkei Stock Average to the outside. The predictive voting processing apparatus 100 controls voting / collecting / refunding. A hypothetical “forecast market” is formed by forecasts of the Nikkei average stock prices of multiple users.

The user registers personal information such as name and account number in the prediction voting processing apparatus 100 before participating in the vote. The user predicts the closing price of the Nikkei Stock Average and votes (bids) his / her predicted value to the prediction voting processing apparatus 100. More specifically, voting information in a predetermined format including the user ID and the predicted value is transmitted from the client terminal 200 to the predicted voting processing apparatus 100. In order to vote, it is necessary to pay consideration (hereinafter referred to as “voting consideration”). In this embodiment, the voting consideration is 1000 yen per unit. When the predictive voting processor 100 receives a vote, it collects the voting consideration from the voter's account. Also, keep track of who and when and how much you voted. The voting consideration is pooled in the management account of the prediction voting processing apparatus 100. Hereinafter, the funds collected in this way are called “original funds”.
The voting acceptance period is from 0:00 to 14:45, and the closing price of the Nikkei Stock Average is fixed at 15:00 after the voting acceptance period expires.

When the closing price is determined, the prediction voting processing apparatus 100 classifies the votes received during the voting reception period into “safe vote” and “out vote”. A safe vote is a vote including an expected value close to the closing price, and a dividend is paid to a voter of the safe vote (hereinafter referred to as “safe voter”). Of course, the voting consideration will also be refunded 100%. The out vote is a vote including an expected value far from the closing price, and no dividend is paid to the voter of the out vote (hereinafter referred to as “out voter”). All or part of the voting consideration will be confiscated. The classification method of the safe vote and the out vote will be described later. The profit / loss of each voter is determined not by whether the Nikkei Stock Average goes up or down, but by how accurately the closing price of the Nikkei Stock Average can be predicted.
In this embodiment, it is assumed that money is sent and received via a bank account. However, as a modification, electronic money settlement means such as “Suica (registered trademark)” and “Osaifu-Keitai (registered trademark)” Money may be sent and received via.

FIG. 2 is a functional block diagram of the prediction voting processing apparatus 100.
Each block shown here can be realized in hardware by an element such as a CPU of a computer or a mechanical device, and in software it is realized by a computer program or the like. Draw functional blocks. Therefore, those skilled in the art will understand that these functional blocks can be realized in various forms by a combination of hardware and software.

The prediction voting processing apparatus 100 includes an external IF (interface) unit 110, a data processing unit 130, and a data holding unit 140.
The external IF unit 110 takes charge of the user interface via the client terminal 200. It is also responsible for interfacing with the bank system 210 and the stock exchange system 220. The data processing unit 130 executes various data processing based on data acquired from the external IF unit 110 and the data holding unit 140. The data processing unit 130 also serves as an interface between the external IF unit 110 and the data holding unit 140. The data holding unit 140 is a storage area for holding various data.

External IF unit 110:
The external IF unit 110 includes a reception unit 112 that receives various data from the client terminal 200 and the like, and a transmission unit 122 that transmits data to the client terminal 200. The receiving unit 112 includes an economic index receiving unit 114, a voting receiving unit 116, a cancellation receiving unit 118, and a price collection unit 120. The economic index receiving unit 114 receives Nikkei average stock price information that changes every moment from the stock exchange system 220. The vote receiving unit 116 receives vote information from the client terminal 200. The voting information includes a user ID, an expected value, and a number. The cancellation receiving unit 118 receives, from the client terminal 200, cancellation information for instructing cancellation of a vote once made. The price collection unit 120 accesses the voter's account and withdraws the voting price. The devoted voting money is pooled as original funds in a dedicated account.

  The transmission unit 122 includes an expected value notification unit 124 and a remittance unit 126. The predicted value notification unit 124 notifies each client terminal 200 of the trend of the predicted value. For example, the maximum predicted value, the variation of the predicted value, the number of votes, etc. may be notified. The remittance unit 126 remits dividends from the management account to the safe voter's account. In addition, all or part of the voting fee will be refunded.

Data holding unit 140:
The data holding unit 140 includes a vote holding unit 142, a frequency information holding unit 144, and a setting information holding unit 146. The voting holding unit 142 holds voting information and the voting time in association with each other. The frequency information holding unit 144 holds frequency information. The frequency information is information indicating the frequency distribution of expected values, and will be described in detail with reference to FIGS. The setting information holding unit 146 holds various setting information. Details will be described with reference to FIG.

Data processing unit 130:
The data processing unit 130 includes a vote classification unit 132, a payout control unit 134, and a frequency distribution calculation unit 136. The vote classification unit 132 classifies the accepted vote into a safe vote and an out vote by a method described later. The refund control unit 134 calculates the amount or dividend to be paid back to each voter by a method described later, and causes the remittance unit 126 to perform the refund. The frequency distribution calculation unit 136 obtains a frequency distribution of expected values from the accepted vote group by a method described later.

FIG. 3 is a schematic diagram for explaining the relationship between the transition of the Nikkei average stock price and voting.
This figure shows data for August 28, the horizontal axis represents the time axis, and the vertical axis represents the Nikkei Stock Average. Further, it is assumed that the Nikkei average closing price on August 27 was 12752 yen. Voting acceptance will begin at 0:00 on August 28th. The user of each client terminal 200 votes the predicted value for the Nikkei Stock Average closing price on August 28 to the prediction voting processing device 100 after the start of voting.

The voting acceptance period is divided into seven periods A to G.
A period: 0: 00 to 8:30 (outside the trading hours)
Period B: 8:30 to 9:00 (outside trading hours)
C period: 9: 00 to 11: 00 (within trading hours, prior play)
D period: 11: 00 to 12: 00 (outside trading hours)
E period: 12: 00 to 13: 00 (partly within trading hours)
F period: 13: 00 to 14:00 (within trading hours, listing)
G period: 14: 00 to 14: 45 (within trading hours)
The closing price of the Nikkei Stock Average is fixed at 15:00, which is the end time of the listing.

  The ballot acceptance period and transaction time partially overlap, and users can vote while watching the trend of the Nikkei Stock Average. Therefore, the G period that is close to the closing price is the time zone that is most easily predicted. For this reason, it becomes easy to concentrate voting at the end of the G period, particularly at the end of the voting acceptance period. Therefore, the predictive voting processing apparatus 100 sets incentives for encouraging voting at an early stage such as the A period and the B period. Details will be described later. Voters can also cancel their vote after voting. In case of cancellation, the entire voting fee will be refunded. A part may be used as a cancellation fee for operating funds of the prediction voting processing apparatus 100.

  When voting in period A, the user predicts the closing price based on the closing price on the previous day, the trend of the Nikkei average stock price so far, news, and the like. In the case of the figure, the expected values accepted in the period A are distributed in the range of 12520 to 13100 yen (maximum width of 580 yen). Expected values accepted in the period B are distributed in the range of 12480 yen to 12960 yen (maximum width 480 yen). There is less variation than period A, and a consensus of voters is being formed somewhat. The timing at which the period C starts is the start timing of the previous stage. The Nikkei Stock Average on the 28th started at the opening price of 12684 yen (68 yen lower than the previous day).

  During the period C, the Nikkei Stock Average has risen after falling. The expected value accepted in the period C is affected by such up and down fluctuations in the Nikkei 225 stock price and is spread to 12450 yen to 12980 yen (maximum width 530 yen). Various forecast values are accepted in each period while being affected by the price movement of the Nikkei Stock Average. The expected value accepted in the G period just before the end of the transaction time is 12820 yen to 12938 yen (118 yen), and the range of the predicted value is the narrowest. It indicates that consensus on the closing price of the voter is becoming more solid because the closing price is close to being confirmed. The voting acceptance period ends at 14:45, and the closing price of the Nikkei Stock Average is fixed at 12910 yen at 15:00.

  After the closing price is settled, votes that are not canceled until the expiration of the voting acceptance period are classified into safe votes and out votes, and payments for payouts and dividends are made. Risks are limited because even out-voters do not lose more than the voting price. On the other hand, safe voters can get a large dividend if they can make an accurate prediction.

FIG. 4 is a diagram illustrating a frequency distribution of predicted values received during the period A.
This figure shows the frequency distribution of votes accepted in the period A on August 28 shown in FIG. The horizontal axis indicates the expected value, and the vertical axis indicates the number of votes. As described with reference to FIG. 3, the expected values received in the period A are distributed in the range of 12520 to 13100 yen (maximum width of 580 yen). Since the material for predicting the Nikkei average stock price is scarce, the variation of the predicted value, that is, the standard deviation σ is also a relatively large value. According to an experiment by the present inventor, many of the votes during the period A are canceled later. If the vote is out without cancelling, the voter loses part or all of the voting price. Therefore, actions such as voting and cancellation are determined while comparing and taking into account the merit of voting at an early stage such as period A and the risk of becoming an out-vote.

FIG. 5 is a diagram illustrating a frequency distribution of predicted values received during the period E.
This figure shows the result of counting the votes accepted in the E period of August 28 shown in FIG. Expected values accepted during period E are distributed between 12605 yen and 12984 yen (maximum width of 379 yen). Since the outlook for the closing price of the Nikkei Stock Average has become clearer than the A period, the standard deviation σ is relatively small.

FIG. 6 is a diagram showing a frequency distribution of expected values accepted over the entire voting acceptance period.
This figure shows the total results of the voting accepted during the entire voting acceptance period of August 28 shown in FIG. Votes that are likely to be out votes are canceled one after another during the voting period, so the expected value will eventually concentrate near the closing price of the Nikkei Stock Average.
If cancellation is prohibited, voting is likely to concentrate during the G period. And since the expected value of G period becomes a value close | similar to a closing price, even if it is a case where cancellation is prohibited, it is thought that an estimated value concentrates on the vicinity of a closing price.

  If a sufficient number of votes can be collected, statistically, the expected value forms a normal distribution, and its median value (mean value) μ is considered to be almost coincident with the closing price. (R1) Votes tend to concentrate near the expiration of the voting acceptance period, and the expected value at the end of expiration is likely to be close to the closing price. (R2) By canceling the voting, the expected value far from the closing price is eliminated. The value μ and the closing price are likely to match.

  In addition, the median μ and the closing price are made closer to each other by eliminating abnormal forecasts, such as extreme forecasts such as the Nikkei Stock Average of 1,000,000 yen, or by adjusting the frequency according to the voting timing. The details of these processes will be described in detail with reference to FIGS.

FIG. 7 is a schematic diagram for explaining the classification criteria for safe votes and out votes.
This figure shows the same distribution as the frequency distribution shown in FIG. 6, but for convenience of explanation, the graph is drawn in the horizontal direction. Let the median be μ and the standard deviation be σ. First, a vote with an expected value that is more than 2σ away from μ is a “lost vote”. Lost vote is a kind of out vote, but all lost vote votes are forfeited. Regardless of the period from A to G, all votes that fall outside the range of μ−2σ to μ + 2σ are lost votes. Hereinafter, the total amount of voting money confiscated for lost votes is called “lost lost funds”.

  Votes with predicted values within μ to nσ (0.7 ≦ n ≦ 1.6) are “safe votes”, and votes with predicted values exceeding nσ are “out votes”. The vote of the expected value of μ ± nσ is zero profit. In other words, μ ± nσ is a breakeven point, and nσ indicates the “threshold value”. If it is larger than nσ, a part of the voting price is confiscated, and if it is further larger than 2σ, the entire voting price is confiscated as a lost vote. On the other hand, if the error between the expected price and the closing price is within nσ, the entire voting price will be refunded, and the dividend will also be paid. The range within μ to nσ, in other words, the range of expected values that are safe votes is called the “safe range”. Hereinafter, the total amount of voting money that is confiscated for out-votes is referred to as “out-source funds”. Lost source funds and out source funds are the source of dividends.

  n is a different value depending on the voting timing. For example, n = 1.04 is set for the voting in the E period, and n = 0.67 is set for the voting in the G period. That is, the voting during the E period has a wider safe range than the voting during the G period. By setting n in this way, the user's willingness to vote at an early stage is aroused.

C1. Method for calculating refund amount for out-vote A part of the 1000 yen voting price is refunded to the out-voter. A method of calculating the refund amount will be described by taking an out vote voted in period E as an example. In the figure, at point B of μ + 1.04σ, the payout amount is 1000 yen, that is, the profit / loss is 0 when the entire voting price is refunded, and at point C of μ + 2σ, the payout amount is 0 yen, ie, the voting price is fully confiscated. It is set that the payout amount decreases in a linear function from point B to point C.
y = a · x + b (y: amount of refund, x: expected value)
age,
1000 = a · (μ + 1.04σ) + b (1)
0 = a · (μ + 2σ) + b (2)
And solving
y = (− 1000 / 0.96σ) · x + {1000 · (μ + 2σ) /0.96σ} ·· (3)
Is obtained.
Here, the refund amount for the expected value of μ + kσ (1.04 ≦ k ≦ 2) is calculated by substituting x = μ + kσ into the above equation (3).
Refund amount y = 1000 · (2-k) /0.96 (4)
As required. Some of the voting money will be refunded, but the rest will be part of the original funds.

C2. Method of calculating dividend points for safe votes A safe voter can receive not only the entire amount of 1000 yen of voting but also dividends. In calculating the dividend, the payout controller 134 first calculates “dividend points”. The higher the dividend points, the higher the dividend. Similarly, the calculation method of the payout point will be described by taking a safe vote voted in the E period as an example. In the figure, μ + 1.04σ (point B) is set to payout point 0, μ (point A) is set to payout point 1, and the payout points are set to decrease linearly from point A to point B.
y = a · x + b (y: dividend point, x: expected value)
age,
1 = a · μ + b (5)
0 = a · (μ + 1.04σ) + b (6)
And solving
y = (− 1 / 1.04σ) · x + (1 + μ / 1.04σ) (7)
Is obtained.
Here, the payout point of the expected value that becomes μ + mσ (0 ≦ m ≦ 1.04) is obtained by substituting x = μ + mσ into the above equation (7).
Dividend points y = (1.04-m) /1.04 (4)
Is obtained. A method for determining the dividend based on the dividend points will be described with reference to FIG.

FIG. 8 is a data structure diagram of the setting information 150.
The setting information 150 is held in the setting information holding unit 146. The voting time column 152 indicates a time zone when voting is performed, the distribution adjustment value column 154 indicates a distribution adjustment value, the frequency adjustment value column 156 indicates a frequency adjustment value, and the safe range column 158 indicates a safe range.

  The distribution adjustment value is a parameter for adjusting the payout point according to the voting timing. For example, in the case of a safe vote calculated with a payout point of 0.6 points, the payout control unit 134 increases the payout point by 10 (1000%) to 6 when the safe vote is cast during the period A. Adjust to point. If the vote is during the period B, the payout point is adjusted to 3 points, which is five times (500%). On the other hand, if the vote is during the G period, the payout point is adjusted to 0.18, which is 0.3 times (30%). Therefore, if a safe vote can be cast at an early stage of the voting reception period, it is possible to expect a large payout point, and thus a large payout.

Here, the relationship between the original funds and dividends is summarized as follows.
Total amount of voting money (original funds) = total amount of refunds to out voters + total amount of refunds to safe voters Total amount of refunds to safe voters = total amount of refunds for safe voters Total amount of dividends to Safe Voters Total amount of dividends to Safe Voters = Lost source funds + Out source funds

The payout control unit 134 calculates a payout for each safe voter by (lost original funds + out original funds) × (dividend points / total payout points). For example, it is assumed that the lost original funds are 1 billion yen, the out original funds are 19 billion yen, and the total payout points of the safe voter is 1 million points. The payout of the safe voter with the payout point = 0.6 is (1 billion yen + 19 billion yen) × (0.6 / 1 million) = 12,000 yen.
The conditions for increasing the dividend are summarized as follows.
S1. Increase your number.
S2. Vote for the expected value close to the closing price S3. Vote early. The higher the total number of votes (original funds), the higher the dividend per safe voter.

  The frequency adjustment value is a parameter for adjusting the frequency distribution shown in FIGS. 6 and 7. A higher frequency adjustment value is set for an expected value that is later in voting timing. When calculating the final frequency distribution shown in FIGS. 6 and 7, the frequency distribution calculation unit 136 adjusts the frequency to 10% for the predicted value accepted in the A period and predicts it in the G period. The value is adjusted to 500%. Usually, the expected value at a later stage such as the F period or the G period is closer to the closing price than the expected value at an earlier stage such as the A period or the B period. Therefore, by adjusting the frequency distribution diagram (histogram) by such weighting, the median μ and the closing price are made closer to each other.

FIG. 9 is a flowchart showing the settlement process after the closing price is determined.
First, when the voting acceptance period expires and the closing price of the Nikkei Stock Average is fixed, the frequency distribution calculation unit 136 calculates the frequency distribution of the expected value (S10). In the frequency distribution diagram, the frequency distribution calculation unit 136 extracts an expected value that is 3σ or more away from μ as an “incorrect predicted value”. The voting classification unit 132 classifies a vote including an expected incorrect value into a lost vote. For example, if an extreme expected value such as 1 yen or 1 million yen is input, the frequency distribution will not accurately reflect the voter's true expectation. Therefore, first, such an incorrect predicted value is extracted, and the frequency distribution calculation unit 136 calculates the frequency distribution again from the predicted value group after the incorrect predicted value is excluded (S12).

  The frequency distribution calculation unit 136 adjusts the frequency distribution by the frequency adjustment value corresponding to the voting time (S14). By the process of S14, the median value μ is usually closer to the closing price, and the standard deviation σ is reduced. The frequency distribution calculation unit 136 calculates the median μ and the standard deviation σ (S16).

  The voting classification unit 132 classifies votes that are not canceled at the end of the voting expiration period and that include normal predicted values that are not incorrect predicted values, into safe votes and out votes. The vote classification unit 132 first specifies a vote to be verified (S18). If the expected value of this vote is within the safe range, it is a safe vote, and if it is outside the safe range, it is an out vote. If it is a safe vote (Y in S20), the payout controller 134 calculates a payout point (S22). In the case of an out vote (N in S20) and not a lost vote (N in S24), the withdrawal control unit 134 calculates a withdrawal amount (S26). In the case of a lost vote (Y in S24), the refund amount is zero.

  If unverified votes remain (N in S28), the process returns to S18. If the classification is completed for all the votes (Y in S28), the payout control unit 134 calculates a payout for each safe voter (S30). The withdrawal control unit 134 causes the remittance unit 126 to execute a withdrawal process (S32). Part of the voting price will be refunded to the out-voter, and the voting price and dividend will be paid to the safe voter.

The prediction voting processing apparatus 100 has been described above based on the embodiment.
According to the predictive voting processing apparatus 100, the profit / loss of each voter is determined by whether or not a value at a predetermined point in time, such as the Nikkei average stock price, which fluctuates during the voting acceptance period, for example, whether the closing price can be accurately predicted. Can do. For example, it is expected to be used as a financial product for hedging risks based on price fluctuations in the Nikkei Stock Average. The earlier the voting, the higher the dividend and the wider the safe range. Therefore, voting can be encouraged from the early stage of the voting reception period. For this reason, the user of each client terminal 200 can know the other party's view of the closing price of the Nikkei Stock Average from an early stage. Knowing the other's view of the closing price before the Japanese market opens is also useful in determining the investment policy for the day. For example, in the period A and period B, when forecast values higher than the closing price of the previous day are gathered, it is estimated that there are many market participants who think that the Nikkei Stock Average on the day will rise. Such estimation can be applied to the strategy.

  In addition, by permitting cancellation of voting, an environment in which voters can easily vote is prepared. Since the withdrawal of a vote that is likely to be an out vote is prompted, the median μ and the closing price are likely to coincide. In order to encourage serious predictions, a part of the voting price, for example, about 50 to 100 yen per unit, may be confiscated as a cancellation fee at the time of cancellation. In this case, a psychological deterrent effect against cancellation works, and if the cancellation fee is added to the original funds, the dividend can be increased.

  Since the predictive voting processing apparatus 100 obtains the median μ and the standard deviation σ from the frequency distribution from which the incorrect predicted value is excluded, it is possible to perform a settlement process for only serious predicted values. Further, by adjusting the frequency distribution by weighting the expected value at the later stage than the expected value at the earlier stage, the median value μ and the closing price can be easily matched.

  In the present embodiment, it has been described that the voting price is 1000 yen per unit, and the frequency changes in proportion to the number of units. As a modified example, for example, when three voters, A to C, vote one by one for the expected value of 13,000 yen, the frequency for the expected value of 13000 yen is counted as 3, and the expected value of 13200 yen is obtained. On the other hand, when one voter D is voting three times, the frequency of the expected value of 13200 yen may be counted as 3 or more. The voter D has invested a lot of money for the expected value of 13200 yen compared to the voters A to C, so the expected value of 13200 yen is a more serious expected value than the voters A to C. This is considered to be an accurate forecast value. By adjusting the frequency according to the amount of money to be input, the confidence and seriousness for each predicted value may be reflected in the frequency distribution.

  When the economic index is out of the predetermined range, for example, when the closing price of the Nikkei Stock Average is 2000 yen or more, or fluctuates 15% or more compared to the closing price of the previous day, the refund control unit 134 invalidates all votes, You may refund all voters. In a situation where an abnormality of the stock exchange system 220 is suspected, it is possible to prevent erroneous settlement by invalidating the transaction itself.

  In addition, the expected value of the Nikkei average stock price may be limited to a predetermined range from the previous day's closing price, for example, within ± 1000 yen, to prevent the illegal expected value from being voted. In this example, the Nikkei Stock Average was targeted as an example of economic indicators, but stocks (individual issues), bonds, interest rates, foreign exchange, real estate (various facilities such as land, condominiums, detached houses, resort facilities), commodities ( Various forecast targets can be set as long as the price fluctuates, such as crude oil, electricity, gas, precious metals, iron, chemical), and emission credits. Although the voting acceptance period in this embodiment is set as about half a day from 0:00 to 14:45, a voting acceptance period may be set from several days to several months and years.

In the present embodiment, the median value μ is regarded as an approximate value of the closing price, and the safe vote and the out vote are classified according to the degree of deviation from the median μ. As a modification, safe votes and out votes may be classified according to the degree of deviation from the closing price itself.
If there is no safe vote, all voting fees may be refunded.

  The present invention has been described based on the embodiments. The embodiments are exemplifications, and it will be understood by those skilled in the art that various modifications can be made to combinations of the respective constituent elements and processing processes, and such modifications are within the scope of the present invention. .

It is a schematic diagram which shows the transaction form in a present Example. It is a functional block diagram of a prediction voting processing apparatus. It is a schematic diagram for demonstrating the transition of a Nikkei average stock price, and a vote. It is a figure which shows the frequency distribution of the expected value received during A period. It is a figure which shows the frequency distribution of the estimated value received during E period. It is a figure which shows the frequency distribution of the expected value received over the whole vote reception period. It is a schematic diagram for demonstrating the classification | category reference | standard of a safe vote and an out vote. It is a data structure figure of setting information. It is a flowchart which shows the adjustment process after final price confirmation.

Explanation of symbols

  100 Predictive Voting Processing Device, 110 External IF Unit, 112 Receiving Unit, 114 Economic Indicator Receiving Unit, 116 Vote Receiving Unit, 118 Cancellation Receiving Unit, 120 Price Collection Unit, 122 Transmitting Unit, 124 Expected Value Notification Unit, 126 Remittance Unit, 130 data processing unit, 132 vote classification unit, 134 payout control unit, 136 frequency distribution calculation unit, 140 data holding unit, 142 vote holding unit, 144 frequency information holding unit, 146 setting information holding unit, 150 setting information, 152 voting time Column, 154 distribution adjustment value column, 156 frequency adjustment value column, 158 safe range column, 200 client terminal, 210 bank system, 220 stock exchange system.

Claims (7)

An economic indicator receiver that receives the current value of economic indicators that change over time during the voting period from an external site;
During the voting reception period, a voting receiver that receives a vote including a predicted value of the economic index at a predetermined time after the end of the voting reception period, from a plurality of client terminals;
Votes accepted during the voting acceptance period after the elapse of the predetermined time point are classified into a safe vote in which the difference between the value of the economic indicator at the predetermined time point and an expected value is within a predetermined profit / loss threshold value, and an out-of-out exceeding the profit / loss threshold value. A voting classification section for classifying into votes;
A refund control unit that distributes dividends to voters who cast the safe vote using a set of money collected from the voter of each client terminal as a source,
When the second safe vote is cast after the first safe vote, the payout control unit determines that the dividend to the voter of the first safe vote is the vote of the second safe vote. A predictive voting processing device that distributes a dividend so as to be higher than a dividend to a person. A cancellation receiving unit for receiving a cancellation instruction for instructing cancellation of voting;
2. The vote classification unit classifies votes that have not been canceled at the end of the vote acceptance period among votes accepted during the vote acceptance period into an out vote and a safe vote. Predictive voting processor.   The voting classification unit is configured so that when the second vote is cast after the first vote, the profit / loss threshold for the first vote is set to be larger than the profit / loss threshold for the second vote. 3. The predictive voting apparatus according to claim 1, wherein each vote is classified into an out vote and a safe vote based on the threshold setting information. A frequency distribution calculation unit that calculates a frequency distribution of expected values for votes received during the voting reception period, further includes,
The voting classification unit regards the median value in the frequency distribution as an approximate value of the economic index at the predetermined time point, and classifies each vote into an out vote and a safe vote according to the degree of deviation from the median value. The predictive voting processing apparatus according to claim 1, wherein The frequency distribution calculation unit calculates a new frequency distribution for the remaining votes excluding the votes outside the predetermined range from the median value in the calculated frequency distribution,
5. The predictive voting apparatus according to claim 4, wherein the voting classification unit classifies each vote into an out vote and a safe vote according to a deviation degree from a median value in the new frequency distribution.   The frequency distribution calculating unit calculates a frequency distribution by setting a frequency higher than that of the first vote for a second vote cast after the first vote. Item 6. The predictive voting processor according to Item 4 or 5. A process of receiving current values of economic indicators that change over time during the voting period from an external site;
During the vote acceptance period, a process of receiving a vote including a predicted value of the economic index at a predetermined time after the end of the vote acceptance period from a plurality of client terminals;
Votes accepted during the voting acceptance period after the elapse of the predetermined time point are classified into a safe vote in which the difference between the value of the economic indicator at the predetermined time point and an expected value is within a predetermined profit / loss threshold value, and an out-of-out exceeding the profit / loss threshold value. Processing to classify the votes,
A process of distributing dividends to voters who cast the safe vote as a source of money collected from the voter of each client terminal;
When the second safe vote is cast later than the first safe vote, the dividend to the voter of the first safe vote is more than the dividend to the voter of the second safe vote. A predictive voting program characterized in that dividends are distributed so as to be higher.

Images