A hybrid bidding transaction method and system based on data linkage

Abstract

The present application relates to the field of bulk commodity electronic transaction calculation, and provides a hybrid bidding transaction method and system based on data linkage, which comprises the following steps: according to the historical behavior data of users in the first transaction mode and the second transaction mode, the transaction credit status of the users in different transaction modes is determined, and real-time credit scores are generated; according to the futures market situation data and the spot market transaction data, the spot reference price is determined; in combination with the futures market situation data, the current basis, the futures, the historical volatility of the spot and the correlation coefficient of the spot and the futures price are determined, and in combination with the transaction quantity in the spot transaction request, the futures contract recommendation information is generated; the spot transaction request is distributed to the first transaction mode or the second transaction mode for bidding matching, wherein the bidding matching priority is dynamically adjusted according to the real-time credit score. The present application effectively associates the spot transaction data with the futures transaction data, and optimizes the spot transaction by using the futures information.

Description

Technical Field

[0001] This invention relates to the field of electronic commodity trading technology, and in particular to a hybrid auction trading method, system, and computer-readable storage medium based on data linkage. Background Technology

[0002] Currently, with the development of internet technology and e-commerce, commodity trading platforms have gradually become an important carrier for online commodity trading. Multi-variety commodity trading platforms can support both parties in completing transaction operations such as confirming the trading object, the quantity of the transaction, and the amount of the transaction online, thereby improving transaction efficiency.

[0003] Existing electronic commodity trading platforms typically include spot trading platforms and futures trading platforms. Spot trading platforms are primarily used for listing, bidding, trading, and delivery of spot commodities, while futures trading platforms are mainly used for trading standardized futures contracts and managing risk. Because spot and futures trading are usually deployed on independent platforms, spot trading platforms typically cannot automatically generate hedging recommendations based on real-time futures market data, nor can they utilize cross-trading model credit information for unified management of bidding rankings.

[0004] In existing technologies, users typically need to log in to spot trading platforms and futures trading platforms separately to complete spot purchases, sales, and hedging operations. The transaction data between different platforms are independent of each other, making it difficult to directly use spot trading behavior for futures trading decisions. Furthermore, it is difficult to promptly feed futures market price information back to the spot trading process, thereby increasing the complexity of user operations and hindering the synergy between spot trading and futures risk management.

[0005] Furthermore, existing trading platforms mostly rank bids based on rules such as price priority and time priority, and different trading modes typically employ independent credit evaluation mechanisms, lacking a cross-trading mode credit linkage mechanism. When a user defaults, frequently cancels orders, or submits abnormal bids in one trading mode, the impact on their credit is difficult to extend to another trading mode in a timely manner, resulting in low utilization of credit evaluation. There is still room for improvement in risk control capabilities during the bidding process.

[0006] Therefore, how to establish a hybrid auction trading method and system that can realize the linkage between spot and futures trading data, support spot trading to assist futures decision-making, optimize spot trading by utilizing futures market information, and realize credit linkage between different trading models has become a technical problem that urgently needs to be solved in this field. Summary of the Invention

[0007] To address these issues, the present invention provides a hybrid auction trading method, system, and readable storage medium based on data linkage, in order to overcome the problems of existing technologies such as the isolation between spot trading and futures trading, the lag in capturing spot-futures arbitrage opportunities, and the low efficiency in formulating hedging strategies.

[0008] To achieve the above objectives, the present invention provides a hybrid auction trading method based on data linkage, comprising: In response to a spot trading request from a first user or a second user, the system obtains the user's historical behavior data in the first and second trading modes, determines the user's trading credit status in different trading modes, and generates a real-time credit score based on the credit correlation between different trading modes. Obtain futures market data from third-party futures trading platforms and spot market transaction data from the platform's historical spot transaction database, and determine the spot reference price based on the spot market transaction data; The current basis, historical volatility of futures and spot prices, and correlation coefficient between spot and futures prices are determined based on the futures market data and the spot reference price. Based on the transaction quantity in the spot transaction request, the basis, the futures, the historical volatility of the spot market, and the correlation coefficient, futures contract recommendation information is generated; The spot trading request is allocated to either the first trading mode or the second trading mode for competitive bidding, wherein the priority of competitive bidding is dynamically adjusted based on the real-time credit score.

[0009] Furthermore, the process of generating real-time credit scores based on the credit correlation between different transaction patterns includes: Obtain transaction fulfillment rate, quote cancellation frequency, abnormal behavior, and transaction size data for the first and second transaction modes respectively; The credit scores for the first transaction model and the second transaction model are determined based on the transaction fulfillment rate, bid cancellation frequency, abnormal behavior, and transaction volume data, respectively. The user's real-time credit score is determined based on the correlation between the credit score of the first transaction mode and the credit score of the second transaction mode.

[0010] Furthermore, the method for determining the abnormal behavior score includes: Acquire order cancellation behavior, abnormal quotation behavior, and related transaction behavior in the first and second transaction modes respectively; Abnormal behavior scores are determined based on the order cancellation behavior, abnormal pricing behavior, and related transaction behavior under different transaction models; The impact value of a user's abnormal credit is determined based on the degree of correlation between abnormal behavior scores under different transaction models.

[0011] Furthermore, the first transaction mode is a reverse bidding mode in which a first user lists multiple second users for public bidding, and the second transaction mode is a matching mode in which the first user and the second user list simultaneously.

[0012] Further, the step of allocating the spot trading request to the first trading mode or the second trading mode for competitive bidding includes: In the reverse bidding mode, the system receives spot purchase requests from the first user and spot price requests from multiple second users who are participating in the bidding. The comprehensive score is determined based on the spot price and credit score of the second user, and multiple second users are ranked from high to low according to the comprehensive score, so that the first user can select the second user who will win the bid. In the matching mode, it receives multiple first-user spot purchase requests and multiple second-user spot sale requests for the same type of spot commodity. Multiple request queues for first users and multiple request queues for second users are constructed. When the highest price of the first user is greater than or equal to the lowest price of the second user, all first users and second users with price intersections are identified and prioritized according to the comprehensive score obtained from the price in each request and the credit score of the corresponding user, so as to obtain the matching result set and the corresponding recommended transaction price.

[0013] Furthermore, the method also includes: The user's bidding matching priority in the first and second transaction modes is updated synchronously based on the real-time credit score. The user's credit rating is determined based on the correspondence between the real-time credit score and the preset credit rating threshold. Based on the user's credit rating, corresponding bidding control strategies are implemented, including: restricting transaction permissions, reducing the weight of the credit score in the overall score, or maintaining the default credit score weight.

[0014] Furthermore, the process of generating futures contract recommendation information includes: The hedging ratio is determined based on the correlation coefficient between spot and futures prices, spot price volatility, and futures price volatility. The recommended number of futures contracts is determined based on the hedging ratio and the number of transactions in the spot transaction request. The recommended futures direction is determined based on the transaction direction in the spot transaction request. The recommendation reason and urgency judgment are generated based on the comparison between the current basis and the historical basis average.

[0015] Furthermore, the method also includes: Based on the current futures main contract price and historical basis statistical range in the futures market data, determine the spot price range to be pushed to the corresponding first or second user.

[0016] Furthermore, the present invention also provides a hybrid auction trading system based on data linkage, comprising: The credit scoring module is used to respond to spot trading requests from the first user or the second user, obtain the user's historical behavior data in the first and second trading modes, determine the user's trading credit status in different trading modes, and generate a real-time credit score based on the credit correlation between different trading modes. The market data acquisition module is used to acquire futures market data from third-party futures trading platforms and spot market trading data from the platform's historical spot trading database, and to determine the spot reference price based on the spot market trading data. The parameter calculation module is used to determine the current basis, historical volatility of futures and spot prices, and the correlation coefficient between spot and futures prices based on the futures market data and the spot reference price. The futures contract recommendation module is used to generate futures contract recommendation information based on the transaction quantity in the spot transaction request, the basis, the futures, the historical volatility of the spot market, and the correlation coefficient. The matching module is used to allocate the spot transaction request to a first transaction mode or a second transaction mode for competitive matching, wherein the priority of competitive matching is dynamically adjusted according to the real-time credit score.

[0017] Furthermore, the present invention also provides a computer-readable storage medium storing a plurality of program codes adapted to be loaded and run by a processor to execute the data-linked hybrid auction trading method.

[0018] Compared with existing technologies, the beneficial effects of this invention are as follows: By establishing a data linkage relationship between spot transaction data and futures market data, this invention generates corresponding futures contract recommendation information based on spot transaction requests, thereby enabling real-time assistance from futures market information to spot transaction decisions, improving the ability to link spot and futures market decisions and the efficiency of hedging strategy formulation; simultaneously, it establishes a credit linkage mechanism between different trading modes, uniformly generating real-time credit scores from historical behavioral data under different trading modes, and dynamically adjusting the bidding matching priority in different trading modes based on the real-time credit scores, so that the credit status in one trading mode can simultaneously affect another trading mode, achieving unified credit management across trading modes, improving the reliability of bidding matching, and reducing transaction credit risk. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the data-linked hybrid auction trading method in this embodiment; Figure 2This is a schematic diagram illustrating the process of generating futures contract recommendation information using the data-linked hybrid auction trading method in this embodiment. Figure 3 This is a schematic diagram of the structure of the hybrid auction trading system based on data linkage in this embodiment. Detailed Implementation

[0020] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0021] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0022] To achieve the above objectives, please refer to Figure 1 and Figure 2 This invention proposes a hybrid auction trading method based on data linkage, the method comprising the following steps: Step S100: In response to the spot trading request of the first user or the second user, obtain the user's historical behavior data in the first trading mode and the second trading mode, determine the user's trading credit status in different trading modes respectively, and generate a real-time credit score based on the credit correlation between different trading modes.

[0023] In step S100 of this embodiment of the invention, the historical behavior data includes: transaction fulfillment rate, bid cancellation frequency, abnormal behavior, and transaction amount or scale. The first transaction mode is a reverse bidding mode where a first user lists multiple second users for public bidding, and the second transaction mode is a matching mode where the first user and the second user list simultaneously. First, the transaction credit status under the corresponding transaction mode is determined based on the historical behavior data in the first and second transaction modes, respectively. Then, a real-time credit score for the user is generated based on the credit correlation between the two transaction modes, so that the transaction behavior in one transaction mode can affect the credit evaluation result in the other transaction mode.

[0024] Step S200: Obtain futures market data from a third-party futures trading platform and spot market transaction data from the platform's historical spot transaction database, and calculate the spot reference price based on the spot market transaction data.

[0025] In step S200 of this embodiment, futures market data can be obtained through a third-party futures trading platform, and spot market trading data can be obtained from the historical spot trading database of this platform. The third-party futures trading platform can be a futures company trading system connected to a futures exchange. The futures market data includes contract information, transaction prices, trading volumes, open interest, and candlestick data published during trading hours. In this embodiment, the latest transaction price, closing price, and settlement price of the futures contract are preferably used as the analysis data. The spot market trading data includes information such as trading time, commodity type, trading quantity, transaction price, buyers and sellers, delivery location, and delivery time from historical spot trading records. The spot reference price can be calculated based on the spot trading records of this platform within a preset statistical period, for example, using the volume-weighted average transaction price; in another embodiment, the publicly published commodity price from a third-party market can also be used as the spot reference price.

[0026] Step S300: Based on futures market data and spot reference prices, determine the current basis, historical volatility of futures and spot prices, and the correlation coefficient between spot and futures prices.

[0027] Wherein, the current basis = the previous day's spot reference price - the current real-time futures price or the previous day's closing price; historical volatility is the standard deviation of the spot or futures returns over the past N trading days; the correlation coefficient between spot and futures prices is the correlation coefficient between the spot price series and the futures price series over the past N trading days, and the correlation between spot and futures prices can be calculated using the Pearson correlation coefficient.

[0028] Step S400: Generate futures contract recommendation information based on the transaction quantity in the spot transaction request, the basis, the historical volatility of the spot market, and the correlation parameters between the spot and futures prices.

[0029] In step S400 of this embodiment of the invention, the recommended futures contract information includes: recommended direction, recommended quantity, and recommended contract month.

[0030] Step S500: Allocate the spot transaction request to the first transaction mode or the second transaction mode for bidding matching, wherein the priority of bidding matching is dynamically adjusted according to the real-time credit score.

[0031] In the method of this invention, the spot market can be commodities or forward contracts traded in the spot market. Futures market data and spot market transaction data are obtained through futures trading companies. In traditional trading platforms, both parties in a spot trading platform focus on the physical attributes of the commodity, such as its quality, quantity, delivery location, and delivery time. Futures trading platforms, on the other hand, deal with standardized financial contracts, with the underlying asset being futures contracts. Both parties focus on financial attributes such as contract month, open interest, and margin levels. The trading rules, pricing mechanisms, settlement methods, and risk monitoring logic of these two platforms differ fundamentally, forcing users engaged in commodity trading to switch between two independent systems for a long time. This invention links spot and futures trading data, automatically generating futures contract recommendations based on current spot trading. Users can then complete corresponding futures contract transactions through a third-party futures trading platform, achieving coordinated interaction between spot and futures trading. Spot trading data can be used to generate futures contract recommendations, and real-time futures market price information can be fed back into the spot trading decision-making process, enabling more timely identification of spot-futures arbitrage opportunities. This allows for the linked analysis of spot trading data and futures market data, generating corresponding futures contract recommendations during spot trading, improving the ability to make decisions in a spot-futures linkage, increasing the efficiency of hedging strategy formulation, and reducing users' cross-platform operation costs. Addressing the issue that existing matching transactions primarily rely on price and time priority for sorting, lacking cross-trading mode credit constraints, this invention incorporates historical behavioral data from different trading modes into the dynamic adjustment of bidding matching priorities. This allows abnormal behavior in one trading mode to influence the bidding matching priority in another, achieving cross-trading mode credit linkage and reducing credit risk.

[0032] The specific contents of steps S100 to S500 are described below.

[0033] In step S100 of this embodiment, the process of generating a real-time credit score based on the credit correlation between different transaction modes includes: Step S110: Obtain the transaction fulfillment rate, quote cancellation frequency, abnormal behavior, and transaction size data for the first and second transaction modes, respectively.

[0034] Step S120: Determine the credit score of the first transaction mode based on the transaction fulfillment rate, quote cancellation frequency, abnormal behavior and transaction size data in the first transaction mode; determine the credit score of the second transaction mode based on the transaction fulfillment rate, quote cancellation frequency, abnormal behavior and transaction size data in the second transaction mode.

[0035] Step S130: Generate a real-time credit score based on the credit correlation between the credit score of the first transaction mode and the credit score of the second transaction mode.

[0036] In one embodiment, a preset correlation coefficient can be used to correlate and fuse the credit scores of the first transaction mode and the credit scores of the second transaction mode. When the credit score in either transaction mode changes, the real-time credit score is updated synchronously based on the credit correlation relationship.

[0037] In one embodiment, a weighted method can be used to calculate the credit score of the first transaction mode and the credit score of the second transaction mode. The credit score of each transaction mode can be expressed as: Credit score = ω1 × Transaction fulfillment rate score + ω2 × (100 - Quote cancellation frequency score) + ω3 × (100 - Abnormal behavior score) + ω4 × Transaction size score, where ω1 to ω4 represent the weights of the transaction fulfillment rate score, quote cancellation frequency score, abnormal behavior score and transaction size score, respectively. Each weight can be configured according to the importance of each credit evaluation indicator in different transaction modes, preferably satisfying ω1 > ω3 > ω2 > ω4. The fulfillment rate represents the proportion of historical orders that have actually been settled, and its corresponding score is the fulfillment rate multiplied by 100. The bid cancellation frequency represents the number of bid cancellations within the last thirty days; points are deducted when the number of bid cancellations exceeds a preset frequency threshold of five times, and its corresponding score is min(number of cancellations / frequency threshold × 100, 100). Abnormal behavior represents records of behavior intercepted by abnormal bidding detection within a preset time period M', and its corresponding score is determined based on the order cancellation rate, abnormal group correlation, and abnormal bid frequency within the most recent preset time period M'. Transaction size represents the ratio of cumulative transaction amount to the platform's highest cumulative transaction amount, and its corresponding score is 100 × transaction size. After obtaining the credit scores for the first and second transaction modes respectively, a real-time credit score is generated based on the credit correlation between the two transaction modes, ensuring that historical behavioral data in either transaction mode can influence the final real-time credit score.

[0038] In one specific embodiment, both the first and second transaction mode credit scores are calculated using transaction fulfillment rate, bid cancellation frequency, abnormal behavior, and transaction size scores, with each dimension ranging from 0 to 100 points. The transaction fulfillment rate score carries the highest weight, followed by the abnormal behavior score, then the bid cancellation frequency score, with the transaction size score serving as a supplementary scoring item. A real-time credit score is generated based on the credit correlation between the first and second transaction mode credit scores, ranging from 0 to 100 points. When the real-time credit score is below or equal to 60 points, the credit weight in the overall score is reduced; when the real-time credit score is below or equal to 40 points, the user's trading eligibility is restricted, suspending bidding permissions in the first transaction mode and order placement permissions in the second transaction mode.

[0039] The score for the abnormal behavior in step S120 is obtained using the following steps S121 and S122, specifically including: Step S121: Obtain order cancellation behavior, abnormal quotation behavior, and related transaction behavior in the first transaction mode and the second transaction mode respectively.

[0040] Step S122: Determine the abnormal behavior score for the first transaction mode and the abnormal behavior score for the second transaction mode based on the order cancellation behavior, abnormal quotation behavior, and related transaction behavior.

[0041] Step S123: Determine the user's abnormal credit impact value based on the correlation between the abnormal behavior score of the first transaction mode and the abnormal behavior score of the second transaction mode, and update the real-time credit score based on the abnormal credit impact value.

[0042] More specifically, the abnormal behavior score is determined using a cumulative abnormality score, which is as follows: S coreanomaly =100×min(ΣS coreevent / θ sum ,1); Where, ΣS coreevent θ is the sum of the current exception scores for all blocked requests made by the user within the most recent preset time M'. sum This is the cumulative anomaly score threshold.

[0043] Current Abnormal Score S coreevent The determination is based on a comprehensive assessment of cancellation rates, the degree of association with abnormal groups, and abnormal pricing practices, namely: S coreevent =f(R withdraw S similarity I freq ); Among them, R withdraw S is the order cancellation rate within the most recent preset time M'. similarity I represents the degree of correlation between users and anomalous groups. freq As an indicator of abnormal pricing, I is triggered when the number of bids within a unit of time exceeds a preset threshold. freq Select 1 if the value is 1, otherwise select 0. When S coreevent When the anomaly threshold θ is exceeded, the current request is determined to be an abnormal request and is blocked from entering the bidding matching queue; when the cumulative anomaly score reaches the cumulative anomaly score threshold θ... sum When the abnormal behavior score is updated under the corresponding transaction mode, the abnormal credit impact value is determined based on the correlation between the abnormal behavior score of the first transaction mode and the abnormal behavior score of the second transaction mode.

[0044] A credit association is established between the first and second transaction modes. When anomaly detection and interception, breach of contract, or abnormal bidding behavior occurs in either transaction mode, the transaction credit status in the corresponding transaction mode is updated first. Then, a real-time credit score is generated based on the credit association between the two transaction modes, and the transaction credit status in the other transaction mode is updated simultaneously. This ensures that abnormal behavior in either transaction mode can affect the bidding matching priority in the other transaction mode, thus achieving cross-transaction mode credit linkage.

[0045] The cancellation rate is the frequency with which users actively cancel their bidding requests within the most recent preset time M'. withdraw =N withdraw / (N order +N withdraw ), where R withdraw N represents the order cancellation rate of users within the most recent preset time M'; withdraw N represents the number of bidding requests that users have voluntarily withdrawn within the most recent preset time M'. order This represents the number of bidding requests initiated by users within the most recent preset time M', including completed, cancelled, and expired bidding requests. When the cancellation rate R... withdraw When the value exceeds a preset threshold, the corresponding bidding request is marked as abnormal. The preset threshold can be configured based on the platform's historical transaction data, for example, 0.3.

[0046] A Graph Convolutional Network (GCN) is used to identify anomalous groups. First, a heterogeneous graph is constructed, with each user as a node and the relationships between users as edges. These relationships include device association, IP association, financial association, and behavioral association. Device associations include multiple users logging in with the same device ID; IP associations include multiple users initiating requests from the same IP address; financial associations include multiple users having transfer relationships between their accounts; and behavioral associations include multiple users having highly similar pricing patterns. Then, a group detection algorithm is used to identify potential anomalous groups. Specifically, the subgraph structure is identified using the maximum modularity method; the label propagation algorithm (LPA) is used for semi-supervised propagation of the subgraph, propagating the anomalous labels of the initial seed nodes to their neighbors; finally, the GCN aggregates neighbor information to learn node embedding vectors, mapping users to a low-dimensional space. If the embedding vectors of multiple users are highly similar, they are identified as a suspicious group.

[0047] The user's correlation with the anomalous group is S. similarity = cosine similarity (e i , e j ) = (e i ·e j ) / (||e i || × ||e j ||), where ei and e j These are the embedding vectors obtained by learning the user and any user node in the anomalous group through a graph convolutional network (GCN), respectively. If there exists S... similarity If the correlation coefficient is greater than 0.85, then the two users are determined to belong to the same suspicious group, and the correlation between the user and the abnormal group is S. similarity The embedded vector update period is T hours, where T can be 6 hours or 24 hours, and can be adjusted according to the actual application scenario.

[0048] Quotation Anomaly Indicator I freq The value is 1 if the number of quotes per unit time exceeds the threshold, and 0 otherwise. Credit can be automatically restored if performance is good by setting a preset time M'.

[0049] In step S300 of this embodiment of the invention, the historical volatility of futures or spot markets is calculated using the same formula, and the volatility... Where µ is the average of the spot or futures yield over the most recent preset time period N days from the current date, N is the preset time period, and R is the average of the spot or futures yields. t Let R be the return on day t of the spot or futures market, where t∈N, R t =ln(P) t / P t-1 ), where P t This represents the spot reference price or futures closing price on day t. Higher volatility in both futures and spot markets indicates greater price risk and a stronger urgency to hedge against this risk; conversely, lower volatility indicates relatively stable prices and a lower urgency to hedge in the short term.

[0050] refer to Figure 2 In step S400 of this embodiment of the invention, spot trading generates futures contract recommendation information using the following steps S410 and S420.

[0051] Step S410: Based on the correlation coefficient ρ between spot and futures prices and the spot price volatility σ S and futures price volatility σ f Determine the hedging ratio h; Step S420: Based on the hedging ratio h and the transaction quantity Q in the spot transaction request. spot Determine the recommended number of futures contracts Q futures Based on the trading direction in the spot trading request, the recommended futures direction is determined, and the recommendation reason and urgency judgment are generated based on the comparison between the current basis and the historical basis average.

[0052] It should be noted that recommended futures contracts include one or more of the following: futures direction, quantity, contract type, rationale, and urgency level. Recommended contracts are typically those whose delivery month is closest to the spot delivery date. The hedging ratio h = ρ × (σ)s / σ f ), where h is the optimal hedging ratio; ρ is the correlation coefficient between spot and futures prices; σ s σ represents the historical volatility of the spot price. f This represents the historical volatility of futures prices. The hedging ratio is calculated using the most recent N' trading days, where N' = 60 or N' = 120. This daily data can be set according to actual conditions, with an update cycle of T' days. T' can be 5 or 10 days and is adjusted based on market changes to balance the stability and timeliness of the calculation results.

[0053] Recommended number of futures contracts Q futures = Q spot ×h / contract unit, where the contract unit is determined according to the contract specifications published by the futures exchange, such as 10 tons per lot for corn and 5 tons per lot for copper. After the user confirms the futures transaction details, a futures trading instruction is generated and sent to the futures company's trading system through API interface with the cooperating futures company.

[0054] In one specific embodiment, step S420 generates a recommendation reason and urgency judgment based on the comparison result between the current basis and the historical basis, including: If the current absolute value of the basis is greater than the sum of the historical basis mean and one standard deviation, the basis is too large, and it is recommended to implement hedging as soon as possible.

[0055] If the current absolute value of the basis is less than or equal to the difference between the historical basis mean and one standard deviation, the basis is too small, and execution can be carried out at an opportune time.

[0056] It should be noted that the historical basis mean can be obtained by averaging the basis over the most recent preset period M days, and the historical basis standard deviation can also be calculated using the basis over the most recent preset period M days. Furthermore, the correlation coefficient and price volatility can be used to generate reasons for futures recommendations and their urgency. For example, if the correlation coefficient ρ is less than the preset coefficient, the correlation between futures and spot prices is weak, and the hedging effect may be unsatisfactory, prompting users to make cautious decisions. If the volatility σ is greater than the sum of the historical volatility mean and one standard deviation of historical volatility, it indicates that market volatility is high and the urgency of hedging is high.

[0057] In step S500 of this embodiment of the invention, the allocation of the spot transaction request to the first transaction mode or the second transaction mode for bidding matching is implemented by the following method, including: In the reverse bidding mode, the system receives spot purchase requests from the first user and spot price requests from multiple second users who are participating in the bidding. The comprehensive score is determined based on the spot price and credit score of the second user, and multiple second users are ranked from high to low according to the comprehensive score, so that the first user can select the second user who will win the bid. In the matching mode, it receives multiple first-user spot purchase requests and multiple second-user spot sale requests for the same type of spot commodity. Based on the quotes in each request, construct multiple first user and multiple second user request queues. When the highest price of the first user is greater than or equal to the lowest price of the second user, identify all first users and second users to be matched with price overlap. Prioritize them according to the comprehensive score obtained from the quotes in each request and the credit scores of the corresponding users, and obtain a set of matching results and the corresponding recommended transaction price.

[0058] In step S500, a first user and a second user of the same type of spot commodity are matched. The first user can view the list of second users matched with them, and also their ranking position among each second user. The second user can view the list of first users matched with them, and also their ranking position among each first user. The first or second user can further communicate their transaction intentions online or offline to confirm the specific transaction. Users can view credit rating results related to the transaction for reference in transaction decisions. In another embodiment of the invention, in the matching mode, when the bid corresponding to the highest comprehensive score of the first user is greater than or equal to the bid corresponding to the lowest comprehensive score of the second user, competitive bidding is performed. In the reverse bidding mode, the first user can view the ranked second users bidding, and after the bidding ends, the second user can view their ranking position in the bidding queue.

[0059] The recommended transaction price for each successfully matched first and second user pair is calculated using the following formula: Recommended Transaction Price P recommend； P recommend =β final ×P buy +(1-β final )×P sell； Among them, P recommend The recommended transaction price; P buy The first user's spot order purchase price; P sell The second user places a spot order for sale. β final = Limit(β base × (1 + (C buy - C sell () / 100), 0.3, 0.7), Among them, C buyFor the first user, real-time credit scoring, C sell For the second user's real-time credit score; Limit represents the limiting function to adjust β. final The weighting is limited to the range of 0.3 to 0.7 to ensure that users with higher credit scores receive appropriate preference in the transaction price calculation, while avoiding excessive weighting that could lead to abnormal deviations in the recommended transaction price. β base =0.5×(1+ (D buy - D sell ) / (D buy + D sell +ε)), Among them, D buy The first user queue depth represents the sum of the total order volume for the highest bid tier and the next N-1 bid tiers; D sell The second user queue depth represents the sum of the cumulative order volume for the lowest bid tier and the following N-1 bid tiers; N is the preset number of tiers, ranging from 1 to 5; ε is a very small positive number to prevent the denominator from being zero, such as 0.001.

[0060] When D buy Equals D sell At that time, β base Equals 0.5; when D buy Greater than D sell At that time, β base If the value is greater than 0.5, the recommended transaction price will be adjusted towards the direction of the first user's bid; when D... buy Less than D sell At that time, β base If the value is less than 0.5, the recommended transaction price should be adjusted towards the direction of the second user's bid.

[0061] The recommended transaction price can provide both users with a price reference, promote transactions, and improve transaction efficiency.

[0062] In this embodiment, when the real-time credit score in any transaction mode changes due to order fulfillment, order cancellation, or abnormal behavior, the bidding matching priority in another transaction mode is adjusted synchronously based on the real-time credit score, so that the first transaction mode and the second transaction mode share the same real-time credit evaluation result, thereby realizing the linkage update of credit status and bidding matching priority between different transaction modes.

[0063] In either the first trading mode or the second trading mode, the following is also included: The user's bidding matching priority in the first and second transaction modes is updated synchronously based on the real-time credit score. The user's credit rating is determined based on the correspondence between the real-time credit score and the preset credit rating threshold. Based on the user's credit rating, corresponding bidding control strategies are implemented, including: restricting transaction permissions, reducing the weight of the credit score in the overall score, or maintaining the default credit score weight.

[0064] It should be noted that when the credit score is lower than or equal to the first preset threshold, the user's trading eligibility is restricted, and their bidding rights in the first trading mode and order placement rights in the second trading mode are suspended; when the credit score is greater than the first preset threshold but lower than or equal to the second preset threshold, the credit weight in the comprehensive score is reduced, and the weight coefficient corresponding to the credit score is adjusted to half of the standard weight coefficient; when the credit score is greater than the second preset threshold, the credit weight in the comprehensive score adopts the standard weight coefficient.

[0065] In one embodiment of the present invention, the method further includes: determining the spot price range to be pushed to the corresponding first user or second user based on the current futures main contract price and the historical basis statistical range in the futures market data.

[0066] Specifically, the reasonable price range for spot purchases by the first user is as follows: [Pf+B−σB,Pf+B+σB]; The reasonable selling price range for spot goods for the second user is: [Pf+B−σB,Pf+B+σB]; Where Pf is the current price of the main futures contract, B is the historical basis mean, and σB is the historical basis standard deviation.

[0067] In this embodiment, futures market data can also be used to assist users in determining spot price ranges, linking futures market price information with spot market prices, and providing a reference for users to formulate hedging strategies and judge spot price levels. For example... Figure 3 As shown in the figure, this invention proposes a hybrid auction trading system based on data linkage, the system comprising: The credit scoring module 100 is used to respond to the spot trading request of the first user or the second user, obtain the user's historical behavior data in the first trading mode and the second trading mode, determine the user's trading credit status in different trading modes respectively, and generate a real-time credit score based on the credit correlation between different trading modes. The market data acquisition module 200 is used to acquire futures market data from third-party futures trading platforms and spot market trading data from the platform's historical spot trading database, and to determine the spot reference price based on the spot market trading data. The parameter calculation module 300 is used to determine the current basis, historical volatility of futures and spot prices, and the correlation coefficient between spot and futures prices based on the futures market data and the spot reference price. The futures contract recommendation module 400 is used to generate futures contract recommendation information based on the transaction quantity in the spot transaction request, the basis, the futures, the historical volatility of the spot market, and the correlation coefficient. The matching module 500 is used to allocate the spot transaction request to a first transaction mode or a second transaction mode for competitive matching, wherein the priority of competitive matching is dynamically adjusted according to the real-time credit score.

[0068] In this embodiment of the invention, the historical behavioral data acquired by the credit scoring module includes transaction fulfillment rate, bid cancellation frequency, abnormal behavior, and transaction amount or scale. The first transaction mode is a reverse bidding mode where a first user lists and multiple second users publicly bid; the second transaction mode is a matching mode where a first user and second users list simultaneously. The credit scoring module executes steps S110-S123, calculating the credit score for the first transaction mode and the credit score for the second transaction mode based on the historical behavioral data, and generating a real-time credit score based on the credit correlation between the two transaction modes. The credit scoring module dynamically updates the real-time credit score based on the abnormal behavior detection results. When the abnormal behavior score exceeds a preset abnormal threshold, the interception module intercepts the corresponding bidding request, preventing it from entering the bidding matching queue. When the user does not exhibit abnormal behavior within a preset time M', the real-time credit score is restored and updated based on normal transaction behavior, achieving dynamic adjustment of the credit status.

[0069] In this embodiment, abnormal behavior is not only used to intercept abnormal bidding requests into the bidding matching queue, but also to update the user's abnormal behavior score and participate in the calculation of real-time credit score.

[0070] The market data acquisition module acquires futures market data from third-party futures trading platforms and spot market trading data from the platform's historical spot trading database, and determines the spot reference price based on the spot market trading data. The parameter calculation module executes step S300 above, calculating the current basis, historical volatility, and the correlation coefficient between spot and futures prices based on the futures market data and the spot reference price. Specifically, the current basis is calculated based on the spot reference price and futures price; historical volatility is calculated based on the price return series within a preset time period; and the correlation coefficient between spot and futures prices is calculated based on the corresponding price series, providing parameter support for subsequent futures contract recommendations.

[0071] The futures contract recommendation module is used to execute the above steps S410 and S420, and generate futures contract recommendation information based on the transaction quantity, current basis, historical volatility and correlation coefficient between spot and futures prices in the spot transaction request. The futures contract recommendation information includes one or more of the following: recommendation direction, recommendation quantity, recommended contract, recommendation reason and urgency level.

[0072] The system in this embodiment also includes a spot price recommendation module, which is used to determine the spot price range to be pushed to the first user or the second user based on the current futures main contract price and the historical basis statistical range, so as to provide a reference for users to determine spot quotations.

[0073] In this embodiment, futures market data and spot transaction data are linked. Futures market price information can be used to assist spot price quotations, and spot transaction data can be used to generate futures contract recommendation information, thereby improving the efficiency of users in formulating hedging strategies and judging spot price levels.

[0074] The system in this embodiment also includes an adjustment module, which is used to adjust the user's transaction permissions and bidding matching priority under different transaction modes based on the real-time credit score. When the real-time credit score is lower than a first preset threshold, the user's transaction qualification is restricted; when the real-time credit score is higher than the first preset threshold but lower than a second preset threshold, the credit weight in the comprehensive score is reduced; when the real-time credit score is higher than the second preset threshold, the standard credit weight is used in the comprehensive score calculation.

[0075] The matching module in this embodiment performs step S500, allocating spot transaction requests to either the first or second trading mode for competitive bidding, and dynamically adjusting the bidding priority based on real-time credit scores. In the matching mode, a recommended transaction price is generated based on the bids from both parties, queue depth, and real-time credit scores, providing a price reference for both parties and improving transaction matching efficiency.

[0076] This invention provides a computer-readable storage medium storing multiple lines of program code, characterized in that the program code is adapted to be loaded and run by a processor to execute the above-described data-linked hybrid auction trading method.

[0077] For ease of explanation, only the parts relevant to the embodiments of this application are shown. For specific technical details not disclosed, please refer to the method section of the embodiments of this invention. The computer-readable storage medium can be a storage device formed by various electronic devices. Optionally, in the embodiments of this application, the computer-readable storage medium is a non-transitory computer-readable storage medium.

[0078] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A hybrid auction trading method based on data linkage, characterized in that, include: In response to a spot trading request from a first user or a second user, the system obtains the user's historical behavior data in the first and second trading modes, determines the user's trading credit status in different trading modes, and generates a real-time credit score based on the credit correlation between different trading modes. Obtain futures market data from third-party futures trading platforms and spot market transaction data from the platform's historical spot transaction database, and determine the spot reference price based on the spot market transaction data; The current basis, historical volatility of futures and spot prices, and correlation coefficient between spot and futures prices are determined based on the futures market data and the spot reference price. Based on the transaction quantity in the spot transaction request, the basis, the futures, the historical volatility of the spot market, and the correlation coefficient, futures contract recommendation information is generated; The spot trading request is allocated to either the first trading mode or the second trading mode for competitive bidding, wherein the priority of competitive bidding is dynamically adjusted based on the real-time credit score.

2. The method according to claim 1, characterized in that, The process of generating real-time credit scores based on the credit correlation between different transaction patterns includes: Obtain transaction fulfillment rate, quote cancellation frequency, abnormal behavior, and transaction size data for the first and second transaction modes respectively; The credit scores for the first transaction model and the second transaction model are determined based on the transaction fulfillment rate, bid cancellation frequency, abnormal behavior, and transaction volume data, respectively. The user's real-time credit score is determined based on the correlation between the credit score of the first transaction mode and the credit score of the second transaction mode.

3. The method according to claim 2, characterized in that, The method for determining the abnormal behavior score includes: Acquire order cancellation behavior, abnormal quotation behavior, and related transaction behavior in the first and second transaction modes respectively; Abnormal behavior scores are determined based on the order cancellation behavior, abnormal pricing behavior, and related transaction behavior under different transaction models; The impact value of a user's abnormal credit is determined based on the degree of correlation between abnormal behavior scores under different transaction models.

4. The method according to any one of claims 1 to 3, characterized in that, The first trading mode is a reverse bidding mode in which a first user lists multiple second users for public bidding, and the second trading mode is a matching mode in which the first user and the second user list at the same time.

5. The method according to claim 4, characterized in that, The step of allocating the spot trading request to the first trading mode or the second trading mode for bidding matching includes: In the reverse bidding mode, the system receives spot purchase requests from the first user and spot price requests from multiple second users who are participating in the bidding. The comprehensive score is determined based on the spot price and credit score of the second user, and multiple second users are ranked from high to low according to the comprehensive score, so that the first user can select the second user who will win the bid. In the matching mode, it receives multiple first-user spot purchase requests and multiple second-user spot sale requests for the same type of spot commodity. Multiple request queues for first users and multiple request queues for second users are constructed. When the highest price of the first user is greater than or equal to the lowest price of the second user, all first users and second users with price intersections are identified and prioritized according to the comprehensive score obtained from the price in each request and the credit score of the corresponding user, so as to obtain the matching result set and the corresponding recommended transaction price.

6. The method according to claim 5, characterized in that, Also includes: The user's bidding matching priority in the first and second transaction modes is updated synchronously based on the real-time credit score. The user's credit rating is determined based on the correspondence between the real-time credit score and the preset credit rating threshold. Based on the user's credit rating, corresponding bidding control strategies are implemented, including: restricting transaction permissions, reducing the weight of the credit score in the overall score, or maintaining the default credit score weight.

7. The method according to any one of claims 1 to 4, characterized in that, The process of generating futures contract recommendation information includes: The hedging ratio is determined based on the correlation coefficient between spot and futures prices, spot price volatility, and futures price volatility. The recommended number of futures contracts is determined based on the hedging ratio and the number of transactions in the spot transaction request. The recommended futures direction is determined based on the transaction direction in the spot transaction request. The recommendation reason and urgency judgment are generated based on the comparison between the current basis and the historical basis average.

8. The method according to any one of claims 1 to 4, characterized in that, Also includes: Based on the current futures main contract price and historical basis statistical range in the futures market data, determine the spot price range to be pushed to the corresponding first or second user.

9. A hybrid auction trading system based on data linkage, characterized in that, include: The credit scoring module is used to respond to spot trading requests from the first user or the second user, obtain the user's historical behavior data in the first and second trading modes, determine the user's trading credit status in different trading modes, and generate a real-time credit score based on the credit correlation between different trading modes. The market data acquisition module is used to acquire futures market data from third-party futures trading platforms and spot market trading data from the platform's historical spot trading database, and to determine the spot reference price based on the spot market trading data. The parameter calculation module is used to determine the current basis, historical volatility of futures and spot prices, and the correlation coefficient between spot and futures prices based on the futures market data and the spot reference price. The futures contract recommendation module is used to generate futures contract recommendation information based on the transaction quantity in the spot transaction request, the basis, the futures, the historical volatility of the spot market, and the correlation coefficient. The matching module is used to allocate the spot transaction request to a first transaction mode or a second transaction mode for competitive matching, wherein the priority of competitive matching is dynamically adjusted according to the real-time credit score.

10. A computer-readable storage medium storing a plurality of program codes, characterized in that, The program code is adapted to be loaded and run by a processor to perform the data-linked hybrid auction trading method according to any one of claims 1 to 8.

Images