Decision generation method and device and computer readable storage medium

Abstract

The embodiment of the invention provides a decision generation method, a decision generation device, electronic equipment, a chip, a computer readable storage medium and a computer program product. The method comprises the steps of collecting first data; the first data comprises structured data and unstructured data; the structured data comprises real-time data of a first stock market and real-time data of a first option market; the unstructured data comprises text data in the financial field; obtaining a plurality of parameters corresponding to the current scene according to the first data; the plurality of parameters represent the market condition of the current scene; matching in a historical scene knowledge base according to the plurality of parameters to determine a first historical scene; the first historical scene is a historical scene similar to a current scene; and determining an option investment strategy in the current scene according to the option investment strategy template of the first historical scene.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of financial technology, and specifically to a decision-making method, a decision-making device, an electronic device, a computer-readable storage medium, and a computer program product. Background Art

[0002] In the financial derivatives market, especially in options trading, investors need to quickly obtain market information to formulate risk management and arbitrage strategies. However, traditional financial information services rely primarily on manual analysis and use a single data source. This leads to inaccurate interpretations of market conditions and a long report generation cycle, making it difficult to cope with high-frequency trading scenarios in the options market. Summary of the Invention

[0003] Embodiments of the present application provide a decision-making method, a decision-making device, an electronic device, a chip, a computer-readable storage medium, and a computer program product.

[0004] The decision-making method provided in the embodiment of the present application includes:

[0005] Collecting first data; the first data includes structured data and unstructured data; the structured data includes: real-time data of a first stock market and real-time data of a first options market; the unstructured data includes: text data in the financial field;

[0006] Obtaining, based on the first data, a plurality of parameters corresponding to the current scenario; the plurality of parameters representing the market situation of the current scenario;

[0007] According to the multiple parameters, matching is performed in a historical scene knowledge base to determine a first historical scene; the first historical scene is a historical scene similar to the current scene;

[0008] Determine the option investment strategy for the current scenario based on the option investment strategy template for the first historical scenario.

[0009] The decision-making device provided in the embodiment of the present application includes:

[0010] Multimodal processing module: used to collect first data; the first data includes structured data and unstructured data; the structured data includes: real-time data of a first stock market and real-time data of a first options market; the unstructured data includes: text data in the financial field;

[0011] Option market analysis module: used to obtain multiple parameters corresponding to the current scenario based on the first data; the multiple parameters represent the market situation of the current scenario;

[0012] Option strategy generation module: used to match the multiple parameters in the historical scenario knowledge base to determine a first historical scenario; the first historical scenario is a historical scenario similar to the current scenario;

[0013] The option strategy generation module is used to determine the option investment strategy in the current scenario based on the option investment strategy template of the first historical scenario.

[0014] The electronic device provided in an embodiment of the present application includes: a processor and a memory, the memory is used to store a computer program, and the processor is used to call and run the computer program stored in the memory to execute any decision-making method provided in the embodiment of the present application.

[0015] The chip provided in the embodiment of the present application includes: a processor for calling and running a computer program from a memory, so that a device equipped with the chip executes any decision-making method provided in the embodiment of the present application.

[0016] The computer-readable storage medium provided in the embodiments of the present application is used to store a computer program, and the computer program enables a computer to execute any decision-making method provided in the embodiments of the present application.

[0017] The computer program product provided in the embodiments of the present application includes a computer program, which, when executed by a processor, implements any decision-making method provided in the embodiments of the present application.

[0018] Through the decision-making method provided in the embodiment of the present application, multi-dimensional data is collected, including structured data and unstructured data. The structured data includes: real-time data of the first stock market and real-time data of the first options market; the unstructured data includes: text data in the financial field. Through multi-dimensional data, multiple parameters are obtained, and the market situation can be accurately analyzed. Historical scenarios similar to the current scenario are matched in the historical scenario knowledge base. According to the option investment strategy template of the similar historical scenario, the option investment strategy for the current scenario is automatically generated to ensure the accuracy and timeliness of the option investment strategy. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0020] Figure 1 A schematic diagram of the implementation flow of the decision-making method provided in the embodiment of the present application;

[0021] Figure 2A schematic diagram of the structure of the decision-making device 200 provided in an embodiment of the present application;

[0022] Figure 3 A schematic diagram of the structure of the decision-making device 300 provided in an embodiment of the present application;

[0023] Figure 4 A schematic structural diagram of an electronic device provided in an embodiment of the present application;

[0024] Figure 5 A schematic structural diagram of the chip provided in an embodiment of the present application. DETAILED DESCRIPTION

[0025] The following will describe the technical solutions in the embodiments of this application in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0026] It should be noted that in the embodiments of the present application, the term "and / or" is merely a description of the association relationship between associated objects, indicating that three relationships can exist. For example, A and / or B can represent three situations: A exists alone, A and B exist at the same time, and B exists alone. In addition, in the embodiments of the present application, the character " / " generally indicates that the associated objects are in an "or" relationship.

[0027] In the description of the embodiments of the present application, the term "corresponding" may indicate a direct or indirect correspondence between the two, or an association relationship between the two, or a relationship between indication and being indicated, configuration and being configured, etc.

[0028] To facilitate understanding of the technical solutions of the embodiments of the present application, the relevant technologies of the embodiments of the present application are described below. The following relevant technologies can be arbitrarily combined with the technical solutions of the embodiments of the present application as optional solutions, and they all fall within the protection scope of the embodiments of the present application.

[0029] Introduction to technical terms

[0030] 1) Option: A financial derivative that gives the holder the right, but not the obligation, to buy or sell an underlying asset at a predetermined price within a specified timeframe. Options are categorized as call options and put options, respectively, granting the buyer the right to buy or sell the underlying asset at a strike price before a future date.

[0031] 2) Underlying Asset: The underlying asset referenced in derivative contracts such as options and futures. For options contracts, the underlying asset is typically a financial product such as stocks, stock indices, bonds, or commodities.

[0032] 3) Asset Volatility: Asset volatility refers to the degree of change in asset prices, usually measured by standard deviation (or variance), reflecting the volatility of asset prices. High volatility indicates large price fluctuations, while low volatility indicates small price fluctuations.

[0033] 4) Implied Volatility Surface: The implied volatility surface is a three-dimensional graph that shows the changes in implied volatility of options based on different strike prices and expiration times. Implied volatility refers to the volatility inferred from option prices and plays a crucial role in option pricing models. The implied volatility surface helps understand the market pricing of different options and the market's expectations of future volatility.

[0034] 5) Cross-Market Linkage Effect: Cross-market linkage refers to the mutual influence of price fluctuations and changes across different markets, particularly those in the stock, foreign exchange, and commodity markets. Market linkage means that fluctuations in one market can trigger price changes in other markets, impacting overall market dynamics.

[0035] 6) Structured Data: Structured data refers to data that can be stored in a table, database, or other format with a fixed structure. Structured data is typically quantitative data, such as stock prices, trading volume, and financial statements. It has clear field definitions and a fixed format, making it easy to process and analyze.

[0036] 7) Unstructured Data: Unstructured data, such as text, images, audio, and video, lacks a fixed format, structure, or standard. Typical examples of unstructured data include news articles, social media content, emails, reports, and analytical articles. Unstructured data typically requires data cleaning and processing to extract useful information.

[0037] 8) Volatility Term Structure: The volatility term structure describes how the implied volatility of options with different expiration dates changes over time. The volatility term structure uses the implied volatility of options with different expiration dates to understand the market's expectations of future uncertainty. Typically, shorter-term options have higher volatility, while longer-term options typically have lower volatility, forming a curve.

[0038] 9) Strike Price Distribution: Strike price distribution refers to the distribution of options with different strike prices within an options contract. It reflects investors' expectations of different price levels in the options market and their choices to buy or sell options. Strike price distribution is crucial for market risk management and option pricing.

[0039] 10) Cross-Market Risk Transmission: Cross-market risk transmission refers to the process by which inter-market risks (such as stock market risk, foreign exchange risk, and interest rate risk) are transmitted between different markets. Fluctuations in one market can affect other markets through different channels. For example, fluctuations in global stock markets can affect commodity prices through the foreign exchange market.

[0040] 11) Retrieval-Augmented Generation (RAG): RAG is an AI technique that combines information retrieval and generative models. In RAG, the model first retrieves relevant information from a database through a retrieval phase and then generates the final output based on this information. This technique effectively combines known information with generative capabilities to improve the accuracy and diversity of answers, and is particularly effective in complex tasks.

[0041] 12) Large Language Model (LLM): A large language model (LLM) is a natural language processing model based on deep learning. It learns language patterns and knowledge by processing large amounts of text data. It possesses powerful text generation and comprehension capabilities and can perform a variety of tasks, including text generation, question answering, and text translation. LLMs utilize extensive training data and parameters to effectively understand and generate the complex structure and semantics of natural language.

[0042] In the financial derivatives market, especially in options trading, investors need to quickly access multi-dimensional market information to formulate risk management and arbitrage strategies. However, traditional financial information services rely primarily on manual analysis, which has three core flaws:

[0043] 1) Data fragmentation and lack of timeliness: Existing systems often rely on a single data source (such as market data or text-based research reports) and lack the ability to dynamically integrate structured and unstructured data. Analysis report generation cycles are long, often lagging by several hours, making them difficult to handle in high-frequency trading scenarios.

[0044] 2) Single analysis dimension: Traditional tools focus on basic technical indicators (such as candlestick patterns), and lack the depth of interpretation of option-specific signals (such as volatility term structure, strike price distribution) and cross-market risk transmission, resulting in strategy recommendations that lack adaptability to the characteristics of derivatives.

[0045] 3) Limited level of intelligence: Existing automated analysis is mostly based on rule engines or static models, which cannot achieve complex reasoning. The generation of investment advisory opinions relies on fixed templates and lacks a retrieval and logical verification mechanism for similar historical scenarios, which can easily lead to conclusions that contradict the real-time market.

[0046] Although retrieval-enhanced generation technology has made progress in general fields, its application in financial derivatives scenarios still faces challenges:

[0047] Multimodal data processing bottlenecks: Options analysis requires the simultaneous processing of heterogeneous data, including tables (open interest distribution), charts (volatility surfaces), and text. Traditional RAG systems often use "carving" document parsing (such as optical character recognition (OCR) combined with table recognition), which is inefficient and difficult to preserve logical relationships between data (such as the correlation between the volatility surface and the underlying asset price).

[0048] - Lack of professional knowledge: General RAG systems lack a financial terminology library (such as Put-Call Ratio and Delta Hedging) and derivatives trading rules coding, resulting in frequent errors in basic concepts in the generated content (such as confusing historical volatility with implied volatility).

[0049] - Weak real-time decision support: Existing solutions focus on static knowledge retrieval and fail to implement streaming data access and dynamic strategy optimization. For example, when a black swan event suddenly occurs in another market, the system cannot quickly update analysis conclusions and adjust the arbitrage strategy weights in the current market.

[0050] Figure 1 A schematic diagram of the implementation flow of the decision-making method provided in the embodiment of the present application is shown in FIG. Figure 1 As shown, the method includes the following steps:

[0051] Step 101: Collect first data; the first data includes structured data and unstructured data; the structured data includes: real-time data of a first stock market and real-time data of a first options market; the unstructured data includes: text data in the financial field.

[0052] In an embodiment of the present application, the first data is collected in real time, and the securities trading system is connected through an application programming interface (API) to obtain structured data such as option implied volatility, option holdings, stock indexes, etc. in real time.

[0053] Through the market interface API and option trading system API, set the timer to obtain the real-time data of the first stock market and the first option market at the preset time interval. The stock market data is such as the current price of the index S t , 5-day, 20-day, and 60-day moving averages Index change (based on 60-day moving average) Options market data such as changes in implied volatility Percentile Tilt index Option Term Structure TS IV (T)=σ IV (T)-σ IV (T-1), the difference between the near-month implied volatility and the far-month implied volatility, is used to determine the market's expectation of future volatility, where N is the length of the historical window, l is the indicator function, ΔP is the implied volatility of a 25% out-of-the-money put option; ΔC is the implied volatility of a 25% out-of-the-money call option; σ ATM is the implied volatility of at-the-money options.

[0054] In this embodiment of the application, you can also access other markets simultaneously to obtain data from other markets. For example, if the current market is the A-share market, you can access overseas market data such as US stocks, foreign exchange and commodities to obtain the rise and fall rates. Exchange rate changes and other data.

[0055] In the embodiment of the present application, multi-threaded asynchronous input / output (IO) can be used to perform high-concurrency API requests, and the following calculation formula can be used: Among them, B is the number of data items requested in a single request, R API is the API request rate, T req The time for data request.

[0056] In the embodiment of the present application, for different markets, if the time zones are different, the Coordinated Universal Time (UTC) can be uniformly adopted, and the calculation formula is as follows: T UTC =T Local +ΔT, where T Local is the local time, and ΔT is the time difference between the local time and UTC time.

[0057] In the embodiment of the present application, the collected time can be preprocessed, such as outlier detection and data filling; outliers are filtered using the triple standard deviation method; and data filling is performed using sliding window interpolation. The calculation formula is as follows: t =αx t-1 +(1-α)x t-2 , for example, the value of α can be 0.7.

[0058] Collected data can be stored in MySQL for partitioning and index optimization to improve query efficiency. This ensures data integrity, low latency, and high availability, providing reliable data support for market reviews, options analysis, and cross-market research.

[0059] In the embodiment of this application, for unstructured data, financial text data is extracted from financial texts. For example, key event tags are extracted from text materials such as brokerage morning reports and option strategy research reports. Key event tags include black swan warnings and volatility arbitrage opportunities.

[0060] In the embodiments of the present application, two methods are provided for parsing unstructured text, and the two methods are introduced below.

[0061] The first method is keyword matching + rule extraction. Based on keyword dictionaries and rule matching technology, it accurately extracts key event tags from texts such as brokerage morning reports and option strategy research reports. It is suitable for scenarios with fixed event types and relatively stable structures. The processing flow is as follows:

[0062] Keyword matching

[0063] Text preprocessing process: clean up special symbols, remove HTML tags, line breaks, meaningless spaces, etc.; standardize text, normalize uppercase and lowercase letters, convert full-width / half-width characters, and perform word segmentation.

[0064] Keyword matching process: Construct time keyword library K={K1,K2,…,K n The keyword library K includes multiple keyword sets, each corresponding to an event. For example, K1 corresponds to a black swan event, where K1 = {"extreme volatility", "circuit breaker", "risk aversion"}.

[0065] Text matching process: Match the input text T and obtain the keyword M = {ω|ω∈T,ω∈K}.

[0066] Matching score calculation process: The calculation process is as follows: Among them, S(E i ) is event E i score; l is the indicator function, which takes the value of 1 when ω∈M and 0 otherwise; is the importance of the current keyword in the text T, f(ω,T) is the frequency of ω in the text T, C is the total number of documents in the corpus, n ω is the number of documents containing ω. . .

[0067] Set the threshold τ, if S(E i )>τ, then the key event label of the input text T is determined to be E i ; If multiple key event tags are obtained, the event tag with the highest score is taken as the final keyword matching result.

[0068] Rule extraction

[0069] Use dependency syntax parsing DEP(ω)=dependency_parse(T). If the dependency relationship of ω belongs to {"subject-predicate relationship", "verb-object relationship"}, then ω is an important keyword, ensuring that the extracted keywords are semantically reasonable.

[0070] Pattern matching rules, regular expression matching For example, when the text mentions “market volatility exceeded X%”, it is identified as a volatility-related event.

[0071] The keyword matching + rule extraction method is computationally efficient and suitable for low-latency scenarios, high-precision requirements, and scenarios with relatively stable event categories.

[0072] The second method is to use a pre-trained Transformer + lightweight Prompt. This method performs text parsing based on pre-trained large models (such as ChatGPT, Claude, and DeepSeek). The Prompt design allows the model to directly generate event labels. This method is suitable for scenarios with a large number of event types and dynamic changes. The specific process is as follows:

[0073] Prompt design standardizes the Prompt structure, that is, P = f(T, E), where T is the input text, E is the event category definition, and f is the Prompt generation function, for example:

[0074] Please extract from the following text events that may affect financial markets and categorize them into the following types:

[0075] Black Swan Warning

[0076] Volatility arbitrage opportunities

[0077] Regulatory policy changes

[0078] Others (please specify)

[0079] Text content: '[T]'"

[0080] Control variables, adjust temperature parameter T tempTo control output stability, Among them, T temp The smaller the value, the more stable the output. For example, T temp The value range is 0.3 to 0.5.

[0081] In the large model inference process, call the API to obtain event labels Among them, LLM stands for Large Language Model, It is the event label returned by the large language model. Optimize the large model inference results and calculate the keyword matching confidence If ρ < τ, then return the "other" event label. Calculate the BM25 similarity between the text and the event label like The event label is like The keyword matching + rule extraction method is used as a supplement to improve the accuracy of the results.

[0082] The pre-trained Transformer + lightweight Prompt method does not require model training and directly calls the large model API. It is suitable for scenarios with strong generalization capabilities, dynamically adapts to new events, and has strong scalability.

[0083] Step 102: Based on the first data, multiple parameters corresponding to the current scenario are obtained; the multiple parameters represent the market situation of the current scenario.

[0084] In an embodiment of the present application, the multiple parameters include one or more of the following parameters: implied volatility surface, implied volatility surface slope, option main contract migration, option main contract turnover rate, market sentiment, and option price fluctuations.

[0085] Step 103: According to the multiple parameters, matching is performed in a historical scene knowledge base to determine a first historical scene; the first historical scene is a historical scene similar to the current scene.

[0086] In the embodiment of the present application, the historical scene knowledge base stores historical extreme market cases, such as typical market events such as stock market crashes and violent fluctuations, and stores their key parameters in a structured manner. In the embodiment of the present application, a historical knowledge scene library can also be constructed, and the specific process is as follows:

[0087] Data collection and time window division, extract extreme market conditions from historical market data, and define data time windows. First, define extreme market conditions, and determine whether it is an extreme market window in the following way. If a market period meets two or more of the first conditions, then the market period is determined to be an extreme market time window; the first condition includes: the market decline exceeds the first threshold, the volatility increase exceeds the second threshold, and the liquidity decrease exceeds the third preset threshold; among them, the first threshold can be -7%, the second threshold can be 30%, and the third threshold can be 20%. Determine a complete market evolution time window for each calculated market, such as the warm-up stage: T before the extreme market pre Days (such as 30 days), core event stage: the period when extreme market conditions occur T event (e.g. 5-15 days), recovery phase: the period when the market returns to normal T recover (e.g. 60 days). The final time window W T =[T pre ,T event ,T recover ].

[0088] Volatility surface change path modeling: Calculate the implied volatility surface for different periods and track its dynamic evolution. Volatility surface σ(K,T)=f(S,K,T,r,q,σ ATM ), where: S is the underlying asset price; K is the strike price; T is the expiration time; r is the risk-free interest rate; q is the dividend yield; σ ATM is the implied volatility of the at-the-money option. In order to track the changing path of the volatility surface, the volatility change rate matrix is calculated: Δσ(K,T)=σ t (K,T)-σ t-1 (K, T), where Δσ(K, T)>0 indicates that the implied volatility is rising (market panic). If Δσ(K, T)<0, the implied volatility is falling (market recovery). Measures the slope of the volatility surface If the Skew rises, it indicates that the market's risk aversion is increasing. If the Skew falls, it indicates that market expectations are stabilizing. The volatility surface before, during, and after extreme market conditions can be reduced in dimension using Principal Component Analysis (PCA): Among them, α i is the principal component weight. are the principal component basis vectors.

[0089] Analysis of the migration characteristics of the strike price of the main contract: depicting the changing patterns of option contracts during market fluctuations. In extreme market conditions, the main contract in the market often quickly migrates to deep in the money or deep out of the money: K 迁移 =K 新主力 -K 旧主力, where deep real-valued migration: K 迁移 >0, it indicates that investors are risk-averse. Deep out-of-the-money migration: K 迁移 <0, it indicates that speculation is increasing. Measuring option liquidity under extreme market conditions, if T 换手 If T 换手 A decline indicates that investors have reduced transactions and market liquidity has shrunk.

[0090] Data storage and index optimization: Build an efficient query structure to support retrospective analysis and AI model calls. Each extreme market data in the historical scenario knowledge base is stored as a structured matrix: D(W T )=[σ(K,T),Skew,K 迁移 ,T 换手 ]. The format of each event is as follows:

[0091]

[0092] By partitioning and storing PARTITIONBYRANGE(year) by year and establishing a vector index library, historical data queries D(x,y)=1-cos(x,y) can be accelerated. FAISS or HNSW can be used for efficient retrieval of similar events.

[0093] Based on this, in an optional implementation of the present application, the method further includes: constructing a historical scenario knowledge base; the historical scenario knowledge base includes multiple historical scenarios; the parameters of the historical scenarios include one or more of the following: event name, time window, implied volatility surface, implied volatility surface slope, option main contract migration, option main contract turnover rate; wherein, the time window is the time window of the event at different development stages.

[0094] In an embodiment of the present application, the parameters of the historical scenario also include one or more of the following parameters: changes in volatility surface, strike price distribution, market sentiment, and option price fluctuations.

[0095] The historical scenario knowledge base identifies extreme market conditions through time window partitioning and characterizes market panic based on fluctuations in the volatility surface. It also tracks the shift in strike prices and turnover rates of major contracts to quantify market liquidity characteristics. Data storage utilizes structured matrices, combining partitioned storage with vector indexing to optimize query efficiency, providing high-quality historical data support for option strategy backtesting, market risk assessment, and robo-advisory.

[0096] In an embodiment of the present application, a historical scene knowledge base can be matched based on multiple parameters in the current scene using a similarity algorithm to determine that the historical scene with a similarity exceeding a first preset threshold is the first historical scene.

[0097] Based on this, in an optional implementation manner of the present application, matching in a historical scene knowledge base based on the multiple parameters to determine the first historical scene includes:

[0098] Based on the multiple parameters, similarity is calculated between the current scene and the historical scenes in the historical scene knowledge base using a similarity algorithm, and a historical scene whose similarity exceeds a first preset threshold is determined to be the first historical scene.

[0099] In the embodiment of the present application, multiple parameters can be converted into vector representations in a high-dimensional space, and the similarity between two scenes can be calculated using a similarity algorithm. The similarity algorithm can use a cosine similarity algorithm or a dynamic time warping algorithm. The cosine similarity algorithm calculation formula is as follows: Where X and Y represent the parameter feature vectors of the current and historical scenarios, respectively, and ||X|| and ||Y|| are their moduli. A higher cosine similarity value indicates a higher similarity between the current market and the historical scenarios. The dynamic time warping algorithm is calculated as follows: Among them, X t and Y t Represent the values of the current market and the historical market at time t respectively. The DTW algorithm evaluates the similarity by dynamically aligning these time series.

[0100] In an embodiment of the present application, the RAG engine can be used to first retrieve several historical scenarios that are most similar to the current market scenario from historical data. The parameters of these historical scenarios usually include past market fluctuations, the behavior of the options market, and macroeconomic events of the corresponding period. A deep learning-based retrieval model (such as a Transformer model) is used to perform semantic matching on historical scenarios, comprehensively evaluate market similarity based on multiple parameters, and select historical scenarios with high similarity. The training goal of the model is to be able to identify the historical scenarios that best match the current market conditions by learning a large amount of historical data.

[0101] In the embodiment of the present application, if multiple similar historical scenes are obtained, they can be weighted according to the similarity and importance of each scene. The weighting formula is as follows: Among them, W i is the weight of the i-th historical scene, CosineSimilarity(X,S i ) is the similarity between the current market scenario and the i-th historical scenario.

[0102] Step 104: Determine the option investment strategy for the current scenario based on the option investment strategy template for the first historical scenario.

[0103] In an embodiment of the present application, after determining the first historical scenario, the option investment strategy template is optimized and adjusted according to the option investment strategy template of the first historical scenario and the market conditions in the current scenario to obtain the option investment strategy in the current scenario.

[0104] In this embodiment of the application, the option investment strategy template can be optimized based on current market data, and machine learning algorithms (such as reinforcement learning, Bayesian optimization, etc.) can be used to continuously adjust the parameters of the option investment strategy template to find the best investment strategy. The goal of strategy optimization is to maximize returns and minimize risks. The optimization goal can be expressed as: in, is the expected return of the strategy, is the risk of the strategy, and λ is the risk tolerance parameter.

[0105] In this embodiment, the option investment strategy template is derived by extracting and summarizing successful investment strategies from historical scenarios. The option investment strategy template includes the following information: option buy / sell strategies and option combination strategies. In this embodiment, the generated option investment strategy template for the current scenario may also include risk assessment results, such as "the return volatility of this strategy under similar historical scenarios is 20%."

[0106] Based on this, in an optional implementation of the present application, the option investment strategy includes: an option buying or option selling strategy, an option combination strategy, and a risk assessment result of the option investment strategy.

[0107] In the embodiment of the present application, option investment strategies include: straddle, wide straddle, butterfly and other option combination strategies, which are matched according to the volatility of the current scenario.

[0108] In the embodiment of the present application, strategy iteration can also be performed based on actual market feedback. After the investment strategy is executed, the strategy execution results are collected and optimized through the feedback mechanism to make the generated strategy more in line with market changes.

[0109] In the embodiment of the present application, the market risk level is assessed by tracking the market price fluctuations and the changes in the implied volatility of options in real time. For example, if the change in the implied volatility exceeds a preset threshold (for example, more than 5%), a risk warning is triggered.

[0110] The implied volatility surface is a key factor in option pricing. It reflects the relationship between the implied volatilities of options with different expiration dates and strike prices. Analysis of the option volatility surface helps to assess market risk expectations and sentiment.

[0111] The process of option volatility surface analysis is as follows: Implied volatility surface modeling: Assuming a set of options with different strike prices, the implied volatility surface can be used to model volatility using the strike price (Strike Price, K) and the time to maturity (Time to Maturity, T). The implied volatility IV(K, T) is typically derived using option pricing models (such as the Black-Scholes model). The formula for implied volatility is: Where C(K,T) is the price of an option with a strike of K and an expiration of T, S is the current market price of the underlying asset, and T is the expiration time of the option. Volatility Smile and Volatility Skew: The volatility surface may exhibit a "volatility smile" or "volatility skew" feature. A volatility smile generally indicates that the implied volatility of call and put options is close; while a volatility skew indicates that the market is more sensitive to risk in one direction. Changes in the shape of volatility can reveal changes in market sentiment through standard deviation analysis: VolatilitySmileΔIV(K)=σ high -σ low , where σ high and σ low are the implied volatilities of high and low strike price options, respectively.

[0112] The Put-Call Ratio (PCR) is a commonly used indicator of long-short sentiment in the options market. It measures market sentiment by calculating the ratio of the number of put options purchased to the number of call options purchased during a specific time period. Changes in the PCR value can reveal investors' market sentiment. A high PCR value usually means that the market is bearish, while a low PCR value means that bullish sentiment is dominant. The PCR analysis process is as follows: The calculation formula for PCR is: Among them, N Put is the number of put options purchased, N Call The number of call options purchased. PCR trend analysis: The changing trend of the PCR value can also reveal changes in market sentiment. When the PCR value rises sharply, it usually means that market sentiment is overly pessimistic, and investors are cautious or even bearish about future market trends. Conversely, when the PCR value falls, it indicates that market sentiment is optimistic and bullish sentiment is dominant. Calculating the dynamic changes of PCR: Among them, ΔPCR t is the rate of change of PCR value at a certain time point, PCR t PCR t-1 The PCR values for the current and previous time points, respectively. This calculation helps identify the volatility of the PCR and the intensity of changes in market sentiment.

[0113] Combining the volatility surface and PCR, we can conduct a comprehensive assessment of market sentiment. The following model can be used to quantify risk assessment and market sentiment assessment in the options market:

[0114] The skewness of the implied volatility surface and the PCR value are combined with the Value at Risk (VaR) model to assess the potential risk exposure of the market. VaR is calculated using the following formula: Among them, Z α is the normal distribution quantile corresponding to the confidence level, σ t is the standard deviation (i.e., volatility) of the options market, and T is the investment timeframe. By calculating VaR, we can estimate the maximum possible loss in the options market over a given period of time.

[0115] By combining the shape of the implied volatility surface (such as the volatility smile and skew) with the changing trend of the PCR value, the sentiment state of the options market (such as "bullish," "bearish," or "neutral") and the potential risk level (such as "low risk" or "high risk") are output. For example, if the market price fluctuation exceeds a preset threshold, the market sentiment is judged to be abnormal and a risk warning is issued; if the PCR change value exceeds a preset threshold, the market sentiment is judged to be abnormal and a risk warning is issued; if the implied volatility exceeds a preset threshold, the market sentiment is judged to be abnormal and a risk warning is issued.

[0116] In the embodiment of the present application, a dynamic change analysis report of the options market can also be generated, and the report content includes: the current long and short sentiment of the options market (such as "market sentiment tends to be bullish");

[0117] Morphological analysis of the implied volatility surface (e.g., “volatility is skewed, the market has a higher downside risk”);

[0118] The changing trends of PCR values and their impact on market sentiment; and,

[0119] Option market risk assessment based on VaR model.

[0120] The dynamic change analysis report of the options market is output in the form of a combination of text and charts to provide investors with decision-making references and help investors judge the market sentiment direction and potential market risks.

[0121] In the embodiment of this application, it is also possible to track and analyze the fluctuations of multiple secondary stock markets and evaluate the impact of these market fluctuations on the domestic market. By establishing a quantitative model, it is possible to identify and quantify the cross-market linkage effects, providing a multi-dimensional reference for investment decisions. The analysis process is as follows:

[0122] Cross-market correlation analysis

[0123] In order to evaluate the linkage effect between different markets, it is necessary to analyze the correlation between each market. Pearson correlation coefficient and cointegration analysis can be used. Pearson correlation coefficient is used to measure the linear correlation between two time series. The formula is: Where Cov(X,Y) is the covariance between market X and market Y, σ X and σ Y are the standard deviations of market X and market Y, respectively. The Pearson correlation coefficient ranges from -1 to 1, where values close to 1 indicate a high positive correlation, values close to -1 indicate a high negative correlation, and values close to 0 indicate no correlation.

[0124] Cointegration analysis: It is used to test whether there is a long-term equilibrium relationship between two or more non-stationary time series. The formula is as follows: t =βX t +ε t , where X t and Y t Represent two time series, β is the regression coefficient, ε t The cointegration test can be used to determine whether there is a long-term stable relationship between different markets. The existence of a cointegration relationship can explain the linkage effect between markets.

[0125] Analysis of cross-market transmission effects

[0126] To quantify the transmission effect of different market fluctuations on the domestic market, Granger causality test and Copula model can be used.

[0127] Granger causality test: This is used to test whether the fluctuations in one market have a causal effect on the fluctuations in another market. Suppose there are two markets X and Y. When performing a Granger causality test, the model is as follows: Among them, Y t represents the first stock market (such as the domestic market), X t represents the second stock market (such as the US stock market or the foreign exchange market), α is a constant term, and β i and γ i is the coefficient of the lag period, ε t is the error term. If γ i Significantly different from zero, indicating that X t Y t There is Granger causality.

[0128] Copula model: It is used to quantify the dependency structure between different markets, and is particularly suitable for dealing with nonlinear relationships. The Copula function can accurately describe the joint distribution between different markets by separating the marginal distribution from the dependency structure. Assume that X t and Y tare the return series of the two markets respectively, and the Copula model can be expressed as: C(F X (x),F Y (y)), where C is the Copula function, F X (x) and F Y (y) are the marginal distribution functions of X and Y, respectively. By selecting an appropriate copula model (such as Gaussian copula, t-copula, etc.), the dependence and transmission effects between different markets can be analyzed.

[0129] In the embodiments of this application, a risk transmission model can also be established to assess the potential risks of extreme market fluctuations to the domestic market. For example, by analyzing the volatility of a second stock market (such as the fluctuation of the VIX index) and combining it with the reaction of the domestic market, the intensity and direction of cross-market risk transmission can be quantified.

[0130] Extreme event risk analysis: Use the VaR model to assess the risk of extreme events that are linked across markets and issue warnings when the risk is too high.

[0131] Generate a cross-market linkage analysis report, which includes: correlation analysis results between the multiple secondary stock markets and the first stock market; Granger causality analysis results; and a risk assessment of the impact of market fluctuations in the multiple secondary stock markets on the first stock market. This cross-market linkage analysis report accurately captures linkage effects between different markets and assesses their impact on the domestic market in real time, providing efficient and reliable support for investment decisions.

[0132] Based on this, in an optional embodiment of the present application, the method further includes: monitoring implied volatility, and if the implied volatility exceeds a preset threshold, issuing a risk warning and / or adjusting the option investment strategy; and / or,

[0133] Monitor market sentiment and, if there are abnormal changes in market sentiment, issue risk warnings and / or adjust the options investment strategy; and / or,

[0134] Collecting real-time data from multiple second stock markets and generating a cross-market linkage analysis report based on the real-time data from the multiple second stock markets; the cross-market linkage analysis report includes one or more of the following information:

[0135] Correlation analysis results between the plurality of second stock markets and the first stock market;

[0136] Granger causality analysis results;

[0137] A risk assessment result of the impact of market fluctuations of the multiple second stock markets on the first stock market.

[0138] In the embodiment of the present application, a market trend report can also be automatically generated by analyzing real-time market data to provide the overall market trend and key support points. The implementation process is as follows:

[0139] Data input and preprocessing

[0140] Access real-time stock indices, trading volume, sector rotation data, technical indicators, and other relevant market data through APIs. All input data is cleansed and normalized to ensure consistency and accuracy in subsequent analysis.

[0141] Index rise and fall and trading volume analysis

[0142] The overall market trend is usually predicted and summarized by the relationship between the rise and fall of the stock index and the trading volume. Among them, I t is the index value at the current time point t, I t-1 The index value at the previous time point. Set the volume change rate Among them, V t is the trading volume at the current time point t, V t-1 The index's trading volume is the trading volume at the previous point in time. By analyzing the index's changes in price and trading volume, we can draw basic conclusions about market trends, such as whether the market is in an upward or downward trend and whether trading volume supports the current price trend. If trading volume and index changes in price are consistent (i.e., prices rise and trading volume increases), it indicates a healthy market uptrend. Conversely, if prices rise but trading volume decreases, it may be a "false" rise, indicating market risk.

[0143] Technical indicator analysis

[0144] Combine technical indicators for trend prediction. Commonly used technical indicators include Bollinger Bands, Relative Strength Index (RSI), Moving Average Convergence / Divergence (MACD), etc.

[0145] Bollinger Bands Analysis: Bollinger Bands judge market volatility by calculating the standard deviation of prices. The calculation formula is

[0146] As follows: BBL t =MA t -2σ t , Among them, MA t is the moving average at time point t, σ tis the standard deviation at time point t. If the price breaks through the upper Bollinger Band (BBU) or lower Bollinger Band (BBL), it may mean that the price is about to reverse or the market is overly volatile.

[0147] RSI analysis: RSI is used to determine whether the market is overbought or oversold by calculating the relative strength of prices. The RSI calculation formula is as follows: Among them, RSI t It is the ratio of the average increase to the average decrease over a certain period. An RSI value greater than 70 generally indicates an overbought market, while a value below 30 indicates an oversold market.

[0148] MACD analysis: MACD is an indicator that measures the gap between two exponential moving averages (EMA). The calculation formula is as follows: MACD t =EMA 12 (P t )-EMA 26 (P t ), Signal t =EMA9(MACD t ), where P t The MACD line represents the current price, and EMA stands for Exponential Moving Average. If the MACD line breaks through the Signal line, it indicates a buy signal; otherwise, it indicates a sell signal.

[0149] Sector rotation analysis

[0150] By analyzing the relative strength of different sectors, we can identify the hot sectors in the market and the direction of capital flow. Sector rotation can be analyzed by calculating the Relative Strength Index (RSI) of each sector: Among them, Avg Gain sector,t and Avg Loss sector,t The relative RSI values of the sectors are the average rise and fall of the sectors in a given period. The relative RSI values between sectors can be used to identify the sectors where funds are flowing in, thus helping investors grasp the sector rotation trend in the market.

[0151] Market trend summary and key support point generation

[0152] Support level calculation: Assume that the support level is the lowest point of market reversal, usually the lowest price in the recent period. The calculation formula is S = min (P t-n ,P t-n+1 ,…,P t-1 ), where P tis the price data, and n is the time window.

[0153] Pressure level calculation: The pressure level is usually the highest price in the recent period. The calculation formula is R = max (P t-n ,P t-n+1 ,…,P t-1 ).

[0154] By calculating the current market support and resistance levels, it can provide users with key technical support points for market trends and help investors formulate response strategies during market fluctuations.

[0155] Based on the above analysis, a market trend report is automatically generated, which includes one or more of the following information: overall market trend; key support and resistance level identification; strength and weakness analysis of each sector and capital flow; analysis results of relevant indicators; the relevant indicators include one or more of the following indicators: Bollinger Bands, relative strength index (RSI), and moving average convergence divergence (MACD).

[0156] Based on this, in an optional implementation of the present application, a market trend report is generated based on the first data, and the market trend report is visually displayed; the market trend report includes one or more of the following information:

[0157] Overall market trends;

[0158] Key support and resistance level identification;

[0159] Analysis of the strengths and weaknesses of each sector and capital flows;

[0160] Analysis results of relevant indicators; the relevant indicators may include one or more of the following: Bollinger Bands, Relative Strength Index (RSI), and Moving Average Convergence Divergence (MACD).

[0161] The decision-making method provided in the embodiments of the present application integrates a large amount of information from multiple data sources (such as real-time market data, unstructured text data, and historical knowledge bases) to comprehensively analyze the current situation of the options market and potential market risks. This multi-dimensional analysis method greatly improves the accuracy of market forecasts and avoids the misjudgments and delays that may be caused by a single data source. By acquiring market data in real time and dynamically adjusting strategies, it is possible to respond to market changes in a timely manner, ensuring that investors make efficient and accurate decisions in a rapidly changing market environment. The RAG engine combines historical data and real-time market scenarios to automatically generate timely and targeted options strategies. It can identify historical events similar to current market conditions through dynamic matching and analysis of historical scenarios and generate optimized strategies based on historical data. This intelligent strategy generation method not only improves the accuracy of the strategy, but also ensures that investors can obtain practical and effective investment advice when faced with complex market scenarios. In addition, it also has a real-time risk monitoring mechanism that can automatically trigger risk warnings and adjust strategies when market fluctuations are abnormal, greatly reducing investment risks. Through cross-market linkage analysis, it can track and analyze in real time the impact of fluctuations in other markets on the current market. Through this cross-market analysis, investors can fully grasp the interplay between different markets, optimize their investment strategies, and avoid decision-making errors caused by ignoring external market influences, thereby gaining more comprehensive and forward-looking market insights. Multimodal data fusion technology combines structured and unstructured data, breaking the data type limitations of traditional methods. By analyzing unstructured text such as options market data, financial news, and research reports, it can extract market signals from multiple dimensions, providing investors with a more diverse and comprehensive basis for decision-making. This multimodal data fusion technology not only enriches data sources but also greatly enhances the ability to perceive market sentiment and trend changes, making decisions more accurate and timely. The effective integration of decision execution and risk monitoring ensures accurate execution and real-time adjustment of investment strategies. By automatically executing strategies and monitoring market changes in real time, strategy execution is unaffected by external fluctuations, minimizing risk during execution. Furthermore, a risk warning mechanism promptly identifies unusual market fluctuations and automatically adjusts strategies based on pre-set rules, ensuring that investors can effectively mitigate risks and maintain the security of their funds in diverse market environments.

[0162] Figure 2 A schematic diagram of the structure of the decision-making device 200 provided in an embodiment of the present application is shown in FIG. Figure 2 As shown, the decision-making device 200 provided in the embodiment of the present application includes:

[0163] Multimodal processing module: used to collect first data; the first data includes structured data and unstructured data; the structured data includes: real-time data of a first stock market and real-time data of a first options market; the unstructured data includes: text data in the financial field;

[0164] Option market analysis module: used to obtain multiple parameters corresponding to the current scenario based on the first data; the multiple parameters represent the market situation of the current scenario;

[0165] Option strategy generation module: used to match the multiple parameters in the historical scenario knowledge base to determine a first historical scenario; the first historical scenario is a historical scenario similar to the current scenario;

[0166] The option strategy generation module is used to determine the option investment strategy in the current scenario based on the option investment strategy template of the first historical scenario.

[0167] In an embodiment of the present application, the multimodal processing module is also used to construct a historical scenario knowledge base; the historical scenario knowledge base includes multiple historical scenarios; the parameters of the historical scenarios include one or more of the following: event name, time window, implied volatility surface, implied volatility surface slope, option main contract migration, option main contract turnover rate; wherein, the time window is the time window of the event at different development stages.

[0168] In an embodiment of the present application, the option strategy generation module is used to calculate the similarity between the current scenario and the historical scenarios in the historical scenario knowledge base based on the multiple parameters through a similarity algorithm, and determine that the historical scenario whose similarity exceeds a first preset threshold is the first historical scenario.

[0169] In an embodiment of the present application, the option investment strategy includes: an option buying or selling strategy, an option combination strategy, and a risk assessment result of the option investment strategy.

[0170] In an embodiment of the present application, the decision-making device 200 also includes: a risk monitoring module for monitoring implied volatility, and if the implied volatility exceeds a preset threshold, issuing a risk warning and / or adjusting the option investment strategy; and / or monitoring market sentiment, and if the market sentiment changes abnormally, issuing a risk warning and / or adjusting the option investment strategy; and / or collecting real-time data of multiple second stock markets, and generating a cross-market linkage analysis report based on the real-time data of the multiple second stock markets; the cross-market linkage analysis report includes one or more of the following information: correlation analysis results between the multiple second stock markets and the first stock market; Granger causality analysis results; risk assessment results of the impact of market fluctuations of the multiple second stock markets on the first stock market.

[0171] In an embodiment of the present application, the decision-making generation device 200 also includes: a market review generation module, which is used to generate a market trend report based on the first data and visually display the market trend report; the market trend report includes one or more of the following information: overall market trend; key support and resistance level identification; strength and weakness analysis of each sector and capital flow; analysis results of relevant indicators; the relevant indicators include one or more of the following indicators: Bollinger Bands, relative strength index RSI, and MACD.

[0172] It should be understood by those skilled in the art that Figure 2 The implementation functions of each module in the decision-making device 200 shown can be understood by referring to the relevant description of the aforementioned method. Figure 2 The functions of the modules in the decision-making device 200 shown can be implemented by a program running on a processor, or by a specific logic circuit.

[0173] Figure 3 A schematic diagram of the structure of the decision-making device 300 provided in an embodiment of the present application is shown in FIG. Figure 3 As shown, the decision-making device 300 provided in the embodiment of the present application includes: a multimodal processing module, an intelligent decision support module, and a decision output module; wherein,

[0174] The multimodal processing module is used for real-time data stream access: it connects to the securities trading system through an API interface to obtain structured data such as option implied volatility, option holdings, and stock indices in real time, and simultaneously accesses other market data. This module provides timely and accurate market data through standardized data input. Unstructured text parsing: BERT and TextCNN models are used to perform semantic analysis on unstructured text data such as options market research reports and financial news, extracting key information such as changes in market sentiment and volatility, and converting it into vector form for storage for subsequent analysis. This allows the extraction of valuable market signals from unstructured data. A historical dataset is constructed that includes extreme market conditions such as financial crises and market crashes, storing relevant structured data such as volatility surface changes and strike price distributions. By comparing historical data with current market scenarios, historical scenarios similar to the current market environment can be identified, providing a reference for strategy generation.

[0175] The intelligent decision support module includes a market review generation module, an options market analysis module, and a cross-market linkage analysis module. The market review generation module analyzes real-time market data and, in combination with technical indicators such as index fluctuations, trading volume, and sector rotation, automatically generates market trend reports, helping investors quickly understand the overall market trend and key support points. The options market analysis module uses a combination of option volatility surfaces and PCR analysis to assess long-short sentiment and potential risks in the options market, outputting dynamic analysis reports on the options market and providing real-time investment strategy references. The cross-market linkage analysis module tracks and analyzes fluctuations in other markets and assesses their impact on the current market, helping investors obtain forward-looking cross-market linkage data and providing more comprehensive reference for decision-making. The option strategy generation module uses the RAG engine to retrieve and match historical data, combining the current market scenario with similar historical scenarios to generate targeted option strategy recommendations. It can dynamically adjust the alignment of historical strategy templates with market characteristics to ensure the effectiveness and timeliness of generated strategies.

[0176] The decision execution and risk monitoring module includes: a strategy execution module and a risk monitoring module; the strategy execution module is used to automatically execute the generated option investment strategy based on real-time market data, ensuring that the strategy can be accurately executed according to predetermined rules, and monitor the execution status in real time through a feedback mechanism to ensure that the execution of the strategy is consistent with the market situation; the risk monitoring module is used to monitor market fluctuations, option market sentiment and the impact of external markets in real time. When abnormal market fluctuations occur, a warning is issued through the risk early warning mechanism, and the investment strategy is automatically adjusted according to preset strategy adjustment rules to ensure investment risk control.

[0177] It should be understood by those skilled in the art that Figure 3The implementation functions of each module in the decision-making device 300 shown can be understood by referring to the relevant description of the aforementioned method. Figure 3 The functions of the modules in the decision-making device 300 shown can be implemented by a program running on a processor, or by a specific logic circuit.

[0178] Figure 4 This is a schematic structural diagram of an electronic device 400 provided in an embodiment of the present application. Figure 4 The electronic device 400 shown includes a processor 410, which can call and run a computer program from a memory to implement the method in the embodiment of the present application.

[0179] Alternatively, as Figure 4 As shown, the electronic device 400 may further include a memory 420. The processor 410 may call and execute a computer program from the memory 420 to implement the method in the embodiment of the present application.

[0180] The memory 420 may be a separate device independent of the processor 410 , or may be integrated into the processor 410 .

[0181] Alternatively, as Figure 4 As shown, the electronic device 400 may further include a transceiver 430 , and the processor 410 may control the transceiver 430 to communicate with other devices, specifically, to send information or data to other devices, or to receive information or data sent by other devices.

[0182] The transceiver 430 may include a transmitter and a receiver. The transceiver 430 may further include an antenna, and the number of antennas may be one or more.

[0183] The electronic device 400 can specifically be the decision-making device of the embodiment of the present application, and the electronic device 400 can implement the corresponding processes implemented by the decision-making device in each method of the embodiment of the present application. For the sake of brevity, they will not be repeated here.

[0184] Illustratively, an embodiment of the present application further provides a computer program product, including a computer program, which can be executed by the processor 410 of the electronic device 400 to complete the steps of any of the aforementioned methods.

[0185] Figure 5 It is a schematic structural diagram of the chip of an embodiment of the present application. Figure 5 The chip 500 shown includes a processor 510, which can call and run a computer program from a memory to implement the method in the embodiment of the present application.

[0186] Alternatively, as Figure 5As shown, the chip 500 may further include a memory 520. The processor 510 may call and execute a computer program from the memory 520 to implement the method in the embodiment of the present application.

[0187] The memory 520 may be a separate device independent of the processor 510 , or may be integrated into the processor 510 .

[0188] Optionally, the chip 500 may further include an input interface 530. The processor 510 may control the input interface 530 to communicate with other devices or chips, and specifically, may obtain information or data sent by other devices or chips.

[0189] Optionally, the chip 500 may further include an output interface 540. The processor 510 may control the output interface 540 to communicate with other devices or chips, and specifically, may output information or data to other devices or chips.

[0190] This chip can be applied to the decision-making device in the embodiments of the present application, and the chip can implement the corresponding processes implemented by the decision-making device in each method of the embodiments of the present application. For the sake of brevity, it will not be repeated here.

[0191] It should be understood that the chip mentioned in the embodiments of the present application can also be called a device-level chip, a device chip, a chip device or an on-chip device chip, etc.

[0192] It should be understood that the processor of the embodiments of the present application may be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above method embodiment can be completed by hardware integrated logic circuits in the processor or software instructions. The above processor can be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. The various methods, steps, and logic block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in the embodiments of the present application can be directly implemented as a hardware decoding processor, or can be implemented by a combination of hardware and software modules in the decoding processor. The software module can be located in a storage medium mature in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, etc. The storage medium is located in the memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware.

[0193] It is understood that the memory in the embodiments of the present application may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct RAM bus random access memory (DR RAM). It should be noted that the memory of the apparatus and methods described herein is intended to include, but is not limited to, these and any other suitable types of memory.

[0194] It should be understood that the above-mentioned memories are exemplary but not restrictive. For example, the memories in the embodiments of the present application may also be static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct RAM RAM (DR RAM), etc. In other words, the memories in the embodiments of the present application are intended to include, but are not limited to, these and any other suitable types of memories.

[0195] The present application also provides a computer-readable storage medium for storing a computer program. This computer-readable storage medium can be applied to the decision-making device in the present application, and the computer program causes a computer to execute the corresponding processes implemented by the decision-making device in each method of the present application. For the sake of brevity, these procedures are not further described here.

[0196] Those skilled in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0197] In the several embodiments provided in this application, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the modules is merely a logical function division. In actual implementation, there may be other division methods, such as multiple modules or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or modules, which can be electrical, mechanical or other forms.

[0198] The modules described as separate components may or may not be physically separate, and the components shown as modules may or may not be physical modules, that is, they may be located in one place or distributed across multiple network modules. Some or all of the modules may be selected to achieve the purpose of the present embodiment according to actual needs.

[0199] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0200] If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a decision-making device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0201] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present application should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. A decision making method, characterized in that: The method comprises: Collecting first data; the first data includes structured data and unstructured data; the structured data includes: real-time data of a first stock market and real-time data of a first options market; the unstructured data includes: text data in the financial field; Obtaining, based on the first data, a plurality of parameters corresponding to the current scenario; the plurality of parameters representing the market situation of the current scenario; According to the multiple parameters, matching is performed in a historical scene knowledge base to determine a first historical scene; the first historical scene is a historical scene similar to the current scene; Determine the option investment strategy for the current scenario based on the option investment strategy template for the first historical scenario.

2. The method according to claim 1, characterized in that Also includes: Construct a historical scenario knowledge base; the historical scenario knowledge base includes multiple historical scenarios; the parameters of the historical scenarios include one or more of the following: event name, time window, implied volatility surface, implied volatility surface slope, option main contract migration, option main contract turnover rate; wherein, the time window is the time window of the event at different development stages.

3. The method according to claim 2, characterized in that The matching in a historical scene knowledge base based on the multiple parameters to determine the first historical scene includes: Based on the multiple parameters, similarity is calculated between the current scene and the historical scenes in the historical scene knowledge base using a similarity algorithm, and a historical scene whose similarity exceeds a first preset threshold is determined to be the first historical scene.

4. The method according to claim 1, wherein The option investment strategy includes: option buying or option selling strategy, option combination strategy, and risk assessment results of the option investment strategy.

5. The method according to any one of claims 1 to 4, characterized in that Also includes: Monitoring implied volatility, and issuing risk warnings and / or adjusting the option investment strategy if the implied volatility exceeds a preset threshold; and / or, Monitor market sentiment and, if there are abnormal changes in market sentiment, issue risk warnings and / or adjust the options investment strategy; and / or, Collecting real-time data from multiple second stock markets and generating a cross-market linkage analysis report based on the real-time data from the multiple second stock markets; the cross-market linkage analysis report includes one or more of the following information: Correlation analysis results between the plurality of second stock markets and the first stock market; Granger causality analysis results; A risk assessment result of the impact of market fluctuations of the multiple second stock markets on the first stock market.

6. The method according to any one of claims 1 to 4, characterized in that The method further comprises: Based on the first data, a market trend report is generated and the market trend report is visually displayed; the market trend report includes one or more of the following information: Overall market trends; Key support and resistance level identification; Analysis of the strengths and weaknesses of each sector and capital flows; Analysis results of relevant indicators; the relevant indicators may include one or more of the following: Bollinger Bands, Relative Strength Index (RSI), and Moving Average Convergence Divergence (MACD).

7. A decision-making device, characterized in that: The device comprises: Multimodal processing module: used to collect first data; the first data includes structured data and unstructured data; the structured data includes: real-time data of a first stock market and real-time data of a first options market; the unstructured data includes: text data in the financial field; Option market analysis module: used to obtain multiple parameters corresponding to the current scenario based on the first data; the multiple parameters represent the market situation of the current scenario; Option strategy generation module: used to match the multiple parameters in the historical scenario knowledge base to determine a first historical scenario; the first historical scenario is a historical scenario similar to the current scenario; The option strategy generation module is used to determine the option investment strategy in the current scenario based on the option investment strategy template of the first historical scenario.

8. An electronic device, characterized in that: include: A processor and a memory, the memory being used to store a computer program, the processor being used to call and run the computer program stored in the memory to execute the decision-making method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that Used to store a computer program, wherein the computer program enables a computer to execute the decision making method according to any one of claims 1 to 6.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the computer program implements the decision making method according to any one of claims 1 to 6.

Images