JP2008209987A - Securities transaction execution system - Google Patents

Abstract

[PROBLEMS] To realize an automatic execution system capable of dynamically changing the order quantity in response to market trends and trader behavior.
SOLUTION: Order schedule definition data for defining the relationship between execution rate transitions and transaction times, an algorithm DB 16 for storing order timing, order schedule change conditions, and order schedule definition data each time the order timing arrives Refer to obtain the scheduled execution rate at each order timing, calculate the planned quantity by multiplying it by the total number of orders, calculate the execution quantity so far, and the difference between the planned quantity and the execution quantity is positive In this case, the execution processing unit 14 generates order data for the corresponding quantity and transmits it to the computer 32 of the exchange, and the execution processing unit 14 sets a change condition for the order schedule set in advance based on the market data. Judgment is made whether or not it is satisfied, and if the change condition is satisfied, the predetermined execution rate at the ordering timing is subject to a predetermined increase / decrease, In by multiplying the order total number of records it makes with the scheduled quantity in the order timing.
[Selection] Figure 7

Description

  The present invention relates to a securities transaction automatic execution system, and more particularly, to a system for automatically buying and selling securities such as stocks according to preset conditions.

As an automatic execution system for securities trading, there has been an automatic trading system described in Patent Document 1 so far.
This automatic trading system divides the total trading volume into a plurality of slices, and has a function of automatically executing orders for the market by associating each slice and the trading price with a trading time division frame (for example, every minute). Yes. In addition, by associating the limit price change logic with each time frame in advance, it has a function of changing the limit price according to the market trend (stock price or trading volume).
JP 2003-281369 A

Patent Document 1 claims that by using this automatic trading system, it is possible for a trader to achieve a certain level of performance without constantly monitoring market trends.
However, in the case of this system, the transaction volume is limited to slices set in advance in each division frame, so the transaction volume cannot be automatically adjusted flexibly in response to market trends. For this reason, in reality, traders cannot keep an eye on the market, and it is necessary to judge the advantages and disadvantages according to the market price and volume, and to adjust the transaction volume.
In addition, when the trader executes such a transaction based on his / her own judgment, there is also a problem in that it is not possible to automatically perform book matching for the corresponding number of subsequent slices.

  This invention has been devised to solve the above-mentioned problems of the prior art, and realizes an automatic execution system that can dynamically change the quantity at the time of ordering in response to market trends and trader behavior. It is aimed.

  In order to achieve the above object, the securities transaction automatic execution system according to claim 1 stores ordering schedule definition data that defines a relationship between a transition of an execution rate related to the total number of orders and a lapse of transaction time. Storage means for storing, data for defining the order timing during the transaction time, and a change condition for increasing or decreasing the scheduled execution rate applied to each order timing based on the order schedule definition data. When ordering data including storage means, brand, order type, and total number of orders is entered, the ordering schedule definition data is referenced each time the ordering timing arrives, and the schedule at each ordering timing Means for obtaining an execution rate, means for multiplying this by the total number of orders and calculating a planned quantity at the order timing; and A means for calculating the execution quantity until the note timing, a means for calculating a difference obtained by subtracting the execution quantity from the planned quantity, and if this difference is a positive value, generate order data for the corresponding quantity, A system having means for transmitting to an exchange computer, and satisfying a preset order schedule change condition based on market data transmitted from the exchange computer at each ordering timing described above If the change condition is satisfied, and if the change condition is satisfied, the planned execution rate at the ordering timing is increased or decreased by a predetermined amount, and this is multiplied by the total number of orders to obtain the planned quantity at the ordering timing. And a means.

  The securities transaction automatic execution system according to claim 2 is the system according to claim 1, wherein the ordering schedule change condition is that a predetermined divergence exists between the past average stock price of the subject issue and the current value. In other words, the ordering schedule changing means increases the scheduled execution rate at the ordering timing when the deviation is advantageous in price, and decreases it when it is unfavorable.

  The securities transaction automatic execution system according to claim 3 is the system according to claim 1 or 2, wherein the order schedule change means further executes a scheduled execution at an order timing before or after a predetermined number of times relative to the order timing. The rate is recognized as the planned execution rate at the ordering timing, and is multiplied by the total number of orders to obtain the planned quantity at the ordering timing.

  In the securities transaction automatic execution system according to claim 1, the implementation so far is based on the planned quantity obtained by multiplying the planned execution rate at each order timing specified by the order schedule and the total number of orders entered. A system is provided in which new order data is transmitted to the exchange computer only when the quantity (cumulative order quantity) is subtracted and the obtained value is greater than zero. For this reason, when the execution pace is faster than planned due to some circumstances, new orders can be automatically suppressed. On the other hand, if the execution pace is behind schedule, orders for larger quantities will be placed accordingly.

In addition, if the system determines that a preset change condition has been met based on market data, the system will automatically adjust the planned quantity at each order timing. The order quantity can be flexibly and automatically adjusted according to the situation.
For example, as described in claim 2, when the change condition defines that there is a predetermined difference or more between the past average stock price and the current value, and this is open in a price advantageous direction In order to increase the planned execution rate and increase the planned volume, increase the difference with the execution volume to temporarily activate the transaction, and if it opens in a disadvantageous direction, the planned execution rate will be lowered It is possible to temporarily suppress the order quantity by reducing the quantity and reducing the difference from the implementation quantity.

  The increase or decrease in the scheduled execution rate at each order timing can also be realized by applying the scheduled execution rate at the order timing before or after a predetermined number of times relative to the specific order timing, as described in claim 3.

  FIG. 1 is a block diagram showing a securities transaction automatic execution system 10 according to the present invention, and this system 10 includes a setting registration unit 12, an execution processing unit 14, and an algorithm DB 16.

  The setting registration unit 12 and the execution processing unit 14 are realized by the CPU of the computer executing necessary processes according to the OS and a dedicated application program. The algorithm DB 16 is provided in the hard disk of the computer, and stores order schedule definition data 18, order strategy definition data 20, and restriction item definition data 22.

To this system 10, a PC 26 operated by a trader and a gateway server 28 are connected.
The gateway server 28 is connected to a host computer 32 of a stock exchange via a communication line 30.

When using this system 10, the trader registers the automatic execution algorithm in the algorithm DB 16 via the setting registration unit 12.
Specifically, when the setting registration unit 12 is accessed from the PC 26, a web form for algorithm definition is transmitted and displayed on the Web browser of the PC 26 (not shown).
Traders define multiple algorithms for automatic execution by selecting and entering various data on this Web form. This algorithm is stored in the algorithm DB 16 via the setting registration unit 12.

FIG. 2 is a system diagram showing a specific example of setting items, in which (1) ordering schedule, (2) ordering strategy, and (3) restrictions are registered under a unique algorithm name.
Here, the “ordering schedule” defines the correspondence between the daily trading time on the stock exchange and the execution rate of orders, that is, the pace of execution. Specifically, a volume curve prepared in advance is used. Defined by selecting.
3 to 6 schematically show specific examples of the volume curve.

First, the graph of FIG. 3 depicts a volume curve of “equal execution” in which the execution rate increases in proportion to the passage of transaction time. The time on the horizontal axis of the graph corresponds to 9:00 to 15:00 when the exchange is open, and the execution rate on the vertical axis of the graph corresponds to the percentage of the number of executions (orders) completed in the total number of orders. (The same applies hereinafter).
In the setting example of FIG. 2, this “equal execution” volume curve is defined.

In the graph of FIG. 4, a volume curve of “volume-weighted execution” corresponding to the daytime volume distribution in a predetermined number of days in the past is drawn.
In this case, orders are executed at a relatively fast pace in the vicinity of the front and back markets, and orders are executed at a relatively slow pace in other times, and the order schedule is relatively consistent with the actual trading pattern. It corresponds to.

  The graph of FIG. 5 depicts a volume curve of “order price standard execution” in which the execution rate is increased immediately after the start of trading and the execution pace remains slow toward closing.

  The graph of FIG. 6 depicts a volume curve of “discount standard execution” in which the pace of execution changes at a slow pace immediately after the start of trading, and the execution rate rapidly increases toward closing.

The trader who has set the ordering schedule of the algorithm by selecting a specific volume curve from these volume curves, then defines the volume curve change logic.
The volume curve change logic means a set of conditions and change results for dynamically changing the volume curve based on data from the market (for example, current value data, quotation data, volume data, etc.).
In the setting example of FIG. 2, the following two change logics are defined.
<1> If the current value is 45 bps or more advantageous than the latest 15 minutes VWAP: Advance the volume curve by 3 minutes <2> If the current value is 45 bps or more disadvantageous than the latest 15 minutes VWAP: Delay the volume curve by 3 minutes

The trader then defines an ordering strategy.
This ordering strategy defines the timing, quantity, and price when the system 10 automatically sends order data to the market, and each stage of the transaction: “Zaraba”, “Close”, “Close” , “Other (arbitrary time)” is set.

In the setting example of FIG. 2, the ordering strategy is set as follows for “Zaraba” located between “close” and “close”.
Timing: 1 minute interval Quantity: Planned quantity-Execution quantity Price at the time of order: Limit price-> Best quotation price tracking Price change logic: Not completed for 5 minutes-> The above "Quantity" is the current scheduled execution rate based on volume curve x It means a value obtained by subtracting the “executed quantity” that is the actual ordering performance up to the present time from the “scheduled quantity” calculated by the total number of orders.
In addition, “Order price → Best price tracking” of “Price when placing an order” is a correction data in which the limit price is attached when new order data is sent, and the price is changed to the price when the best price on the board changes. Is generated and transmitted.
Furthermore, “Price change logic” “5 minutes incomplete → market” means that if the above price at the time of ordering cannot be executed for 5 minutes, correction data for changing the price to market is generated and transmitted. Yes.

In addition, the ordering strategy is set as follows for “delays” immediately after the start of trading.
Timing: 1 minute before the front line Quantity: Planned quantity-Implementation quantity Price: Price at the time of ordering: Banking Price change logic: None

Furthermore, an ordering strategy is set as follows for “closing” just before the end of the transaction.
Timing: 1 minute before closing the market Quantity: Planned quantity-Implementation quantity Price: Price at the time of ordering: Market price Change logic: None

Although not set in the setting example of FIG. 2, in this system 10, by setting a predetermined setting for “others” in the ordering strategy, any time point other than “close”, “zalaba”, and “close” Can develop an ordering strategy.
For example, when specific market data is received, or when a specific calculation index exceeds a threshold, special pricing is applicable.

Next, the trader sets the restrictions.
This restriction is logic for restricting the ordering operation by the execution processing unit 14, and the following two are defined in the setting example of FIG.
<1> Upper limit of trading volume: If the share of the market exceeds 30% of the total trading volume of the market: Stops new orders. <2> Upper limit of board quantity: Your limit order exceeds the remaining quantity on the market board. If 5% is reached: stop additional limit orders

Next, automatic execution processing by the system 10 will be described.
FIG. 7 is a conceptual diagram showing the flow of data during automatic execution, and FIGS. 8 and 9 are flowcharts showing the basic processing procedure at that time.
First, the trader accesses the execution processing unit 14 of the system 10 from the PC 26, inputs the order instruction data 40 including the following items through the order instruction form (not shown) displayed on the Web browser, and enters the execution processing unit 14. Send.
(1) Stock code
(2) Order type (Sell / Buy)
(3) Total number of orders
(4) limit price
(5) Algorithm name In addition, the trader sends individual order instructions based on his / her own judgment within the above total number of orders, even after sending the comprehensive order instruction data (order instruction data for one day). Data can be sent to the enforcement processor 14.

  The execution processing unit 14 that has received the comprehensive order instruction data 40 (S10) receives the order schedule definition data 18, order strategy definition data 20, and constraint definition data 22 related to the algorithm for automatic execution specified by the trader. Read from the algorithm DB 16 (S12).

  Next, the execution processing unit 14 receives market data (current value data, quote data, volume data, etc.) 42 transmitted from the host computer 32 of the exchange as needed (S14).

Next, the execution processing unit 14 calculates VWAP for the latest 15 minutes at every ordering timing (every minute) (S16), and a difference of 45 bps (45 / 10,000) or more is opened between this VWAP and the current value. It is determined whether or not (S18).
Here, VWAP is an abbreviation of Volume Weighted Average Price, and means “volume-weighted average price (equivalent to average stock price)” that calculates the weighted average of trading for each price, taking into account the volume for each price. To do.
If there is a difference of 45 bps or more between the VWAP and the current price for the past 15 minutes, it is determined whether it is advantageous or disadvantageous (S20), and the volume curve changing process is performed according to the determination result. This is executed, which will be described in detail later.
Here, it is assumed that there is no difference of 45 bps or more between the VWAP for the latest 15 minutes and the current value, and the description after S22 will be continued. Of course, if the transaction stage is “close”, even if it is “Zaraba”, since there is no data for less than 15 minutes from the start of the transaction and VWAP cannot be calculated, it is naturally skipped to S22.

In S22, the execution processing unit 14 refers to the volume curve, obtains the current scheduled execution rate, and multiplies it by the total number of orders to calculate the planned quantity.
Next, the execution processing unit 14 calculates the execution quantity (cumulative order quantity) up to the present time (S28), and obtains the difference between the planned quantity and the execution quantity (S30).
If this difference is greater than 0 (S32) and does not violate the restrictions on the upper limit of the volume and the upper limit of the plate quantity (S34, S36), the execution processing unit 14 stores the new order data 44 having a quantity corresponding to this difference. It is generated and transmitted to the host computer 32 of the exchange (S38).
For example, when the limit order is not easily executed, the above-mentioned difference is expanded because the execution quantity is low, and the order quantity at the time of new order is increased accordingly.

On the other hand, if the difference is determined to be 0 or minus in S32, it means that an order exceeding the ordering schedule has already been made, so the execution processing unit 14 generates and transmits new order data at the ordering timing. Retain and wait until the next ordering timing.
For example, if the order is executed ahead of schedule due to volume curve change processing, which will be described later, or if an order exceeding the planned execution rate is made based on the judgment of the trader itself, the value of “planned quantity-executed quantity” is It can be 0 or less.
In addition, when it is determined in S34 that the upper limit of the volume is violated and in S36, it is determined that the upper limit of the plate quantity is violated, the execution processing unit 14 generates order data at the ordering timing. Refrain from sending and wait until the next ordering timing.

  If it is determined in S18 that there is a difference of 45 bps or more between the current 15-minute VWAP and the current value, and it is determined in S20 that it is an “advantageous” difference, the execution processing unit 14 A process for advancing the volume curve for 3 minutes is executed according to the volume curve change logic. Here, “advantageous” means “cheap” in the case of a buy order, and “high” in the case of a sell order.

  Specifically, the execution processing unit 14 first refers to the volume curve, obtains the planned execution rate at the order timing after 3 minutes (the order timing after 3 minutes), and multiplies this by the total number of orders. (S24), the execution quantity up to the present time is calculated (S28), the difference between the two is calculated (S30), and if this is greater than 0, the order data corresponding to the difference is generated, It transmits to the host computer of an exchange (S32, S38).

  In this way, the planned quantity is calculated by regarding the planned execution rate after 3 minutes as the current scheduled execution rate, and this is compared with the current implementation quantity. The difference between “scheduled quantity-executed quantity” increases, and it becomes possible to automatically advance the order in a situation advantageous to the user.

  On the other hand, when it is determined in S20 that the difference is “unfavorable”, the execution processing unit 14 executes a process of delaying the volume curve for 3 minutes in accordance with the volume curve change logic. Here, “disadvantage” means “high” in the case of a buy order, and “cheap” in the case of a sell order.

  Specifically, the execution processing unit 14 first refers to the volume curve, obtains the planned execution rate at the ordering timing three minutes before (the ordering timing three orders before), and multiplies this by the total number of orders. After calculating (S26), the execution quantity up to the present time is calculated (S28), the difference between the two is calculated (S30), and if this is 0 or less, the new ordering at the current ordering timing is stopped (S32).

  In this way, the planned quantity is calculated by regarding the planned execution rate 3 minutes ago as the current scheduled execution rate, and this is compared with the current implementation quantity. The possibility that the difference between “scheduled quantity-executed quantity” becomes 0 or less increases, and it becomes possible to automatically suppress additional orders under disadvantageous conditions.

10 and 11 are graphs showing specific examples of volume curve change.
In FIG. 10, a diagram showing the transition of VWAP on the day and a diagram showing the transition of the current value are drawn. Compared with VWAP, the location a where the current value is 45 bps or higher and the location where the current value is over 45 bps. b, c, d, e are shown.
On the other hand, in FIG. 11, a diagram showing the transition of the execution quantity of the purchase order and a diagram showing the volume curve of the equal execution are drawn.
Comparing FIG. 10 and FIG. 11, it can be seen that the execution rate is flat at the high-priced portion a, and the execution rate is abruptly advanced at the low-priced portions b, c, d, and e.

The automatic execution processing of S16 to S38 by the execution processing unit 14 is continued until the trading time of the stock exchange is over (S30).
During this time, monitoring data 60 such as market data and order data is transmitted from the execution processing unit 14 to the trader's PC 26 at any time and displayed on the Web browser.

  In the above, “change the volume curve for 3 minutes” and “delay the volume curve for 3 minutes” are exemplified as the method for changing the volume curve. However, the present invention is not limited to this. It can also be “increase the ratio (for example, 3%)” or “decrease the volume curve by a predetermined ratio (for example, 3%)”.

“Increase the volume curve by 3%” means that when the planned execution rate at a certain order timing is 7% according to the volume curve, this is increased to 10% and the current planned quantity is calculated. This means that as a result, the difference between the planned quantity and the implementation quantity will increase.
“Reduce the volume curve by 3%” means that if the planned execution rate at a certain order timing is 10% according to the volume curve, this is reduced to 7% and the current planned quantity is calculated. As a result, the difference between the planned quantity and the implementation quantity is reduced.

Next, the price setting function of the system 10 will be described.
First, according to the setting example of FIG. 2, “execution” is set as the order price for “close” and “close”, and the price change logic is not set. All that is required is to generate order data for “success” and send it to the host computer 32 of the exchange.
On the other hand, “Zaraba”, which accounts for the majority of the transaction time, is set to “Price → Best Price Tracking” as the order price and “5 minutes incomplete → market” as the price change logic. The execution processing unit 14 executes more complicated processing.
Hereinafter, the price operation of the execution processing unit 14 in “Zaraba” will be described with reference to the flowchart of FIG.

When each order timing arrives, the execution processing unit 14 first generates order data specifying “limit” as a price, and transmits it to the host computer 32 of the exchange (S50).
Next, when the execution processing unit 14 detects that a change has occurred in the best quote on the board based on the market data 42 (S52), correction data for changing the previous “limit” to “best quote” is obtained. It is generated and transmitted to the host computer 32 of the exchange (S54).
Furthermore, when 5 minutes have passed since the order is placed (S56), it is determined whether or not the order related to the order is fulfilled (S58). If the order is not promised, the price is changed from the previous "best bid". Correction data to be changed to “execution” is generated and transmitted to the host computer 32 of the exchange (S60).

  The price setting in FIG. 2 is an example, and “limit” can be set for “close” and “close”. For “Zaraba”, “limit” may be set as the ordering price, the price change logic may not be set, or only “success” may be set.

It is a block diagram which shows the securities transaction automatic execution system which concerns on this invention. It is a systematic diagram which shows the specific example of the setting matter in algorithm DB. It is a graph which shows the volume curve of equal execution. It is a graph which shows the volume curve of volume weighted execution. It is a graph which shows the volume curve of order price standard execution. It is a graph which shows the volume curve of discount standard execution. It is a conceptual diagram which shows the flow of the data at the time of automatic execution. It is a flowchart which shows the process sequence at the time of automatic execution. It is a flowchart which shows the process sequence at the time of automatic execution. It is a graph which shows contrast with a present value and VWAP. It is a graph which shows the specific example of an order schedule change. It is a flowchart which shows the procedure of price operation in Zaraba.

Explanation of symbols

10 Securities transaction automatic execution system
12 Setting registration section
14 Executive Processing Department
16 Algorithm DB
18 Order schedule definition data
20 Order strategy definition data
22 Restriction definition data
28 Gateway server
30 Communication line
32 Exchange host computer
40 Order instruction data
42 Market data

Claims (3)

Storage means for storing order schedule definition data that defines the relationship between the transition of execution rate relative to the total number of orders and the passage of transaction time;
Storage means for storing data defining the timing of ordering during trading hours;
Storage means for storing a change condition for increasing or decreasing a planned execution rate applied to each order timing based on the order schedule definition data;
Means for obtaining a planned execution rate at each order timing by referring to the order schedule definition data each time the order timing arrives when order instruction data including a brand, an order type, and the total number of orders is input;
Multiply this by the total number of orders and calculate the planned quantity at the ordering timing;
Means for calculating the quantity of execution until the ordering timing;
Means for calculating the difference obtained by subtracting the execution quantity from the planned quantity,
When this difference is a positive value, it is a system comprising means for generating order data for a corresponding quantity and transmitting it to a computer of an exchange,
Means for determining whether or not a preset order schedule change condition is satisfied based on market data transmitted from an exchange computer at each ordering timing;
If this change condition is satisfied, an order schedule change means for making a predetermined increase / decrease in the scheduled execution rate at the order timing and multiplying this by the total number of orders to obtain the planned quantity at the order timing;
A securities transaction automatic execution system characterized by comprising: The ordering schedule change condition is that there is a predetermined divergence between the past average stock price and the current value of the subject stock,
2. The securities transaction automatic execution system according to claim 1, wherein the order schedule changing means increases the scheduled execution rate at the order timing when the deviation is advantageous and decreases the disadvantage when the deviation is unfavorable. .   The ordering schedule changing means recognizes the scheduled execution rate at the ordering timing before or after a predetermined number of times relative to the ordering timing as the scheduled execution rate at the ordering timing, and multiplies this by the total number of orders to obtain the planned quantity at the ordering timing. The securities transaction automatic execution system according to claim 1 or 2, characterized in that:

Images