Parameter detection method and device, storage medium and electronic device

Abstract

The embodiment of the invention provides a parameter detection method and device, a storage medium and an electronic device, and the method comprises the steps: obtaining an object parameter of a target object in a target prediction time period from a target prediction platform, and enabling the target object to be any one of a plurality of objects; generating a transaction parameter set of the target object by using the object parameters of the target object, the transaction parameter set including a group of transaction parameters of the target object and a target fluctuation amplitude of the first transaction parameter, and the target fluctuation amplitude being a value greater than a preset amplitude in continuous variation amplitudes of the first transaction parameter in a plurality of historical time periods; performing normal distribution calculation on the transaction parameter set to obtain an object reference threshold value of the target object in the target prediction time period; and performing anomaly detection on the second transaction parameter based on the object reference threshold and the group of transaction parameters to obtain a first anomaly detection result. Through the transaction parameter detection method and device, the technical problem that the transaction parameter detection accuracy is low in the prior art is solved.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of computers, and more specifically, to a parameter detection method, device, storage medium, and electronic device. Background Art

[0002] At present, the relevant technologies mainly rely on manual anomaly detection of predicted transaction parameters. That is, the staff needs to perform anomaly detection on the transaction parameters based on a large amount of historical data and manual experience, which will cause the transaction parameter detection results to be affected by the staff's subjective influence. Different staff members will detect different anomaly results, which makes the monitoring performance of the transaction parameters low and cannot be applied to the actual business system, resulting in the technical problem of low accuracy of transaction parameter detection.

[0003] To address the above-mentioned problems, no effective solution has been proposed yet. Summary of the invention

[0004] The embodiments of the present application provide a parameter detection method, device, storage medium and electronic device to at least solve the technical problem of low detection accuracy of predicted transaction parameters in the related art.

[0005] According to one aspect of the present application, a parameter detection method is provided, which is applied to a data detection device, wherein the data detection device is connected to a target prediction platform, and the target prediction platform is used to predict object parameters of multiple objects based on multiple transaction factors. The method includes: obtaining object parameters of a target object in a target prediction time period from the target prediction platform, wherein the target object is any one of the multiple objects; generating a transaction parameter set of the target object using the object parameters of the target object, wherein the transaction parameter set includes a set of transaction parameters of the target object and a target fluctuation range of a first transaction parameter, and the target fluctuation range is a value greater than a preset range among the continuous change ranges of the first transaction parameter in multiple historical time periods; performing normal distribution calculation on the transaction parameter set to obtain an object reference threshold of the target object in the target prediction time period; performing anomaly detection on a second transaction parameter based on the object reference threshold and the set of transaction parameters to obtain a first anomaly detection result, wherein the second transaction parameter is a parameter generated by the target prediction platform in the process of performing a risk prediction operation on the target object in the target prediction time period.

[0006] According to another aspect of the present application, a parameter detection device is provided, which is applied to a data detection device, wherein the data detection device is connected to a target prediction platform, and the target prediction platform is used to predict object parameters of multiple objects based on multiple transaction factors. The device includes: a first acquisition module, which is used to acquire object parameters of a target object in a target prediction time period from the target prediction platform, wherein the target object is any one of the multiple objects; a first generation module, which is used to generate a transaction parameter set of the target object using the object parameters of the target object, wherein the transaction parameter set includes a set of transaction parameters of the target object and a target fluctuation range of a first transaction parameter, and the target fluctuation range is a value greater than a preset range in the continuous change range of the first transaction parameter in multiple historical time periods; a first calculation module, which is used to perform normal distribution calculation on the transaction parameter set to obtain an object reference threshold of the target object in the target prediction time period; and a first detection module, which is used to perform anomaly detection on a second transaction parameter based on the object reference threshold and the set of transaction parameters to obtain a first anomaly detection result, wherein the second transaction parameter is a parameter generated in the process of the target prediction platform performing a risk prediction operation on the target object in the target prediction time period.

[0007] In an exemplary embodiment, the above-mentioned device also includes: a second acquisition module, which is used to obtain the historical parameter values ​​of the first trading parameter in multiple historical time periods before generating the trading parameter set of the target object using the object parameters of the target object; a first calculation module, which is used to calculate the volatility in each historical time period using the historical parameter values ​​in each historical time period to obtain multiple first volatility; a first determination module, which is used to determine the average value of the multiple first volatility as the first volatility, wherein the historical parameter values ​​in each historical time period have the same calculation weight in the process of calculating the first volatility; a second calculation module, which is used to calculate the second volatility in multiple historical time periods according to the weight corresponding to each historical time period and multiple first volatility; a first input module, which is used to input the historical parameter values ​​in multiple historical time periods into a target statistical model to obtain a third volatility, wherein the target statistical model is a statistical model by analyzing the volatility in time series data; a third calculation module, which is used to calculate the target volatility based on the first volatility, the second volatility and the third volatility.

[0008] In an exemplary embodiment, the above-mentioned first calculation module includes: a first input unit, used to input the set of trading parameters included in the trading parameter set and the target fluctuation range into the normal distribution model to obtain the object reference threshold, wherein the normal distribution model is determined by performing normal distribution modeling on the set of trading parameters and the target fluctuation range.

[0009] In an exemplary embodiment, the normal distribution model calculates the object reference threshold by the following formula: C = K*np.exp(-r*T)*N_d1-F*N_d2; wherein, C is used to represent the object reference threshold, d1 = (np.log(F / K)+(r+0.5*sigma**2)*T) / (sigma*np.sqrt(T)), d2 = d1-sigma*np.sqrt(T), N_d1 = no rm.cdf(d1), N_d2=norm.cdf(d2), K, F, r, and T are all parameters in the set of trading parameters, sigma is used to represent the target volatility, np.log is used to represent the logarithm of F / K, np.sqrt is used to represent the square root of T, and norm.cdf is used to represent the cumulative distribution function of the normal distribution, which is used to calculate the probability that a normally distributed random variable is less than or equal to d1 or d2.

[0010] In an exemplary embodiment, the above-mentioned first detection module includes: a first acquisition unit, used to obtain the second transaction parameter in response to a parameter detection request; a second input unit, used to input the object reference threshold, the set of transaction parameters and the second transaction parameter into a target classification model to obtain a classification result, wherein the target classification model includes a feature extraction module and a classification module, the feature extraction module is used to extract features of the object reference threshold and features of the set of transaction parameters to obtain a first feature description matrix, and is used to extract features of the second transaction parameter to obtain a second feature description matrix, and merge the first feature description matrix and the second feature description matrix to obtain a target feature description matrix, and the classification module is used to perform a classification operation on the target feature description matrix to obtain a classification label; the first detection unit is used to detect the second transaction parameter based on the classification label to obtain the first anomaly detection result.

[0011] In an exemplary embodiment, the above-mentioned device also includes: a fourth calculation module, which is used to detect the second transaction parameter based on the classification label, and after obtaining the first anomaly detection result, when it is determined that the second transaction parameter is abnormal based on the classification label, calculate the first difference between the second transaction parameter and the object reference threshold; a first prediction module, which is used to send the first difference to the target prediction platform, so that the target prediction platform predicts abnormal prompt information based on the abnormality level, wherein the abnormal prompt information includes abnormal description information, abnormal causes, abnormal impacts and abnormal countermeasures, and the abnormal prompt information is also used to instruct the target prediction platform to adjust the transaction factor of the transaction parameter predicted for the target object.

[0012] In an exemplary embodiment, the above-mentioned device also includes: a first updating module, which is used to perform anomaly detection on the second trading parameter based on the object reference threshold and the set of trading parameters, and after obtaining the first anomaly detection result, update the multiple historical time periods according to a preset time period to update the target fluctuation range; a first execution module, which is used to re-execute the normal distribution calculation on the trading parameter set according to the updated target fluctuation range to determine the updated object reference threshold; a second detection module, which is used to re-execute anomaly detection on the second trading parameter based on the updated object reference threshold and the set of trading parameters to obtain a second anomaly detection result; a first comparison module, which is used to compare the first anomaly detection result with the second anomaly detection result to obtain a comparison result; and a first determination module, which is used to determine the target detection result of the second trading parameter based on the comparison result.

[0013] According to another embodiment of the present application, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps in any of the above method embodiments are implemented.

[0014] According to another embodiment of the present application, a computer-readable storage medium is provided, in which a computer program is stored, wherein the computer program is configured to execute the steps of any of the above method embodiments when running.

[0015] According to another embodiment of the present application, an electronic device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0016] In an embodiment of the present application, a normal distribution calculation is performed on a transaction parameter set of a target object to obtain an object reference value, and anomaly detection is performed on a second transaction parameter based on the object reference value and a set of transaction parameters to obtain an anomaly detection result. Since the transaction parameter set includes a set of transaction parameters of the target object and the fluctuation range of the first transaction parameter, the anomaly detection result is more accurate, thereby improving the accuracy of anomaly detection, thereby solving the technical problem of low detection accuracy of predicted transaction parameters in related technologies. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 is a hardware structure block diagram of a mobile terminal according to a parameter detection method of an embodiment of the present application;

[0018] Figure 2 is a flow chart of a parameter detection method according to an embodiment of the present application;

[0019] Figure 3 is a flow chart of a parameter detection method according to an embodiment of the present application;

[0020] Figure 4 is a flow chart of another parameter detection method according to an embodiment of the present application;

[0021] Figure 5 It is a structural block diagram of a parameter detection device according to an embodiment of the present application. DETAILED DESCRIPTION

[0022] The embodiments of the present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0023] It should be noted that the terms "first", "second", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0024] The method embodiments provided in the embodiments of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Taking running on a mobile terminal as an example, Figure 1 FIG. 1 is a hardware structure block diagram of a mobile terminal of a parameter detection method according to an embodiment of the present application. Figure 1 As shown, the mobile terminal may include one or more ( Figure 1 Only one is shown in the figure) a processor 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA) and a memory 104 for storing data, wherein the mobile terminal may also include a transmission device 106 and a terminal device 108 for communication functions. It can be understood by those skilled in the art that Figure 1The structure shown is only for illustration and does not limit the structure of the mobile terminal. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations shown.

[0025] The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the parameter detection method in the embodiment of the present application. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, to implement the above method. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories may be connected to the mobile terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0026] The transmission device 106 is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of the mobile terminal. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0027] Optionally, in this embodiment, the above parameter detection method may also be implemented by a server, or implemented by the terminal device and the server together.

[0028] The above is only an example and is not specifically limited in this embodiment.

[0029] Optionally, as an optional implementation, Figure 2 is a flow chart of a parameter detection method according to an embodiment of the present application, such as Figure 2 As shown, the above parameter detection method is applied to a data detection device, the data detection device is connected to a target prediction platform, and the target prediction platform is used to predict object parameters of multiple objects based on multiple transaction factors. The above method includes:

[0030] S202, obtaining object parameters of a target object in a target prediction time period from the target prediction platform, wherein the target object is any one of the multiple objects;

[0031] Optionally, in this embodiment, it may include but is not limited to logging into the target prediction platform through a front-end device to obtain the object parameters of the target object in the target prediction time period. The front-end device may include but is not limited to a mobile phone, a computer, a tablet, etc., and the target prediction platform may include but is not limited to various website systems, independent application systems, cloud service systems deployed on cloud servers, etc. The above object parameters may include but are not limited to the cost range of an object in a certain area in the next year predicted by the target prediction platform, and the events that will occur in a certain area in the future period.

[0032] Optionally, in an embodiment of the present application, the target object may include but is not limited to foreign exchange spot, forward, swap, option and combination futures of the above products, virtual items, etc. The object parameters correspond to the target object and may include but are not limited to attribute information of the target object, for example, the price of the target object, the transaction fee of the target object, the exchange rate of the target object, etc.

[0033] For example, taking a futures trading platform as an example, the target object represents an option product, and the object parameters may include but are not limited to the exercise price, futures price, annualized expiration time, exchange rate, etc. in a certain region predicted by various websites, systems and other channels in the next year.

[0034] S204, generating a transaction parameter set of the target object using the object parameters of the target object, wherein the transaction parameter set includes a set of transaction parameters of the target object and a target fluctuation range of a first transaction parameter, and the target fluctuation range is a value of the first transaction parameter that is greater than a preset range among continuous fluctuation ranges of the first transaction parameter in multiple historical time periods;

[0035] Optionally, in this embodiment, a group of transaction parameters included in the transaction parameter set are all parameters related to the transaction costs of the target object. For example, the target object is an option product, and a group of transaction parameters includes the futures price, strike price and annualized expiration time of the option product predicted by the target prediction platform for the next year. Among them, the strike price (Strike Price) is the price specified in the option contract, at which the underlying asset can be bought (call option) or sold (put option) on the expiration date. The futures price (Futures Price) is the transaction price of the futures contract of the underlying asset in the current market. The annualized expiration time (t) is the time from the current moment to the expiration date of the option, usually in years. If expressed in other time units (such as days), it needs to be converted into years.

[0036] Exemplarily, the first trading parameter is the price of the option product, which is obtained by obtaining the historical price data of the target object in the past three years from various websites and trading platforms. The volatility of the target object in the past three years is calculated, and the maximum value is selected as the target volatility. Among them, volatility is the annual volatility of the price of the underlying asset, usually expressed as a percentage. This parameter reflects the extent to which the price of the underlying asset may change in the future.

[0037] S206, performing normal distribution calculation on the transaction parameter set to obtain an object reference threshold value of the target object within a target prediction time period;

[0038] Optionally, in an embodiment of the present application, the above-mentioned target prediction time period may include but is not limited to any time period in the future, and the above-mentioned object reference threshold may be understood as the upper limit value or the lower limit value of the second transaction parameter in the target prediction time period.

[0039] Alternatively, the normal distribution is a continuous probability distribution that describes the outcomes of a large number of independent random events.

[0040] S208, performing anomaly detection on a second transaction parameter based on the object reference threshold and the set of transaction parameters to obtain a first anomaly detection result, wherein the second transaction parameter is a parameter generated during a process in which the target prediction platform performs a risk prediction operation on the target object within the target prediction time period;

[0041] Optionally, in the embodiment of the present application, the second transaction parameter may be a parameter in a set of transaction parameters, or may be a parameter corresponding to the transaction of the target object. For example, the target object is an option product, and the second transaction parameter is the option premium of the option product predicted by the target prediction platform.

[0042] For example, the option premium is the fee paid for buying or selling an option contract, which usually consists of two parts: intrinsic value and time value. Option premium = intrinsic value + time value, and the time value depends on factors such as the remaining time to expiration, the volatility of the underlying asset, and the risk-free interest rate.

[0043] Intrinsic value is the profit that can be obtained when the option is exercised immediately. For a call option, if the current price of the underlying asset is higher than the strike price, the intrinsic value is positive, otherwise it is zero. For a put option, if the current price of the underlying asset is lower than the strike price, the intrinsic value is positive, otherwise it is zero.

[0044] For example, suppose there is a call option with an exercise price of 100 yuan and the current price of the underlying asset is 105 yuan. Then the intrinsic value of the call option is 105 yuan (current price) - 100 yuan (exercise price) = 5 yuan. If the current price of the underlying asset is 95 yuan, then the intrinsic value of the call option is zero, because exercising the option will not bring any benefits.

[0045] Time value is the portion of the option premium that exceeds the intrinsic value. It reflects the impact of factors such as the remaining time to expiration of the option, the volatility of the underlying asset, and the risk-free interest rate on the option price. The time value gradually decreases as the expiration date approaches, and on the expiration date, the time value is zero.

[0046] For example, suppose there is a call option with an exercise price of 100 yuan, the current price of the underlying asset is 102 yuan, and the remaining expiration time is 1 year. Assume that the risk-free interest rate is 3% and the volatility of the underlying asset is 20%. In this case, the intrinsic value of the option is 2 yuan (current price-exercise price). However, since the option still has 1 year to expiration, and the underlying asset has a certain volatility and the impact of the risk-free interest rate, the actual price of the option (i.e., the option premium) will be higher than its intrinsic value. This part that is higher than the intrinsic value is the time value. In order to calculate the specific option premium (including intrinsic value and time value), complex mathematical models such as the Black-Scholes Model are required. These models take into account factors such as the current price of the underlying asset, the exercise price, the remaining expiration time, the volatility of the underlying asset, the risk-free interest rate, etc. to estimate the reasonable price of the option.

[0047] Alternatively, the formula for commodity option premium is: Strike Price*NORMSDIST(-(LN(Futures Price / Strike Price)-0.5*Volatility*Volatility*Annualized Time to Expiration t) / (Volatility*SQRT(Annualized Time to Expiration t)))-Futures Price*NORMSDIST(-(LN(Futures Price / Strike Price)+0.5*Volatility*Volatility*Annualized Time to Expiration t) / (Volatility*SQRT(Annualized Time to Expiration t))). Combined with the formula for commodity option premium, it can be seen that strike price, futures price, volatility and annualized time to expiration are all important parameters. These parameters are calculated through specific mathematical functions to obtain the value of the option premium. This formula for commodity option premium is a variant of the Black-Scholes model (usually used to estimate the reasonable price of financial derivatives) and is used to estimate the theoretical price of commodity options. This formula is mainly used to calculate the price of European options (options that can only be exercised on the expiration date).

[0048] In an exemplary embodiment, Figure 3is a flow chart of a parameter detection method according to an embodiment of the present application. The parameter detection method proposed in the present application can be applied to the application scenario of option trading platform for option premium detection, such as Figure 3 This may include, but is not limited to:

[0049] S302, start;

[0050] S304, obtaining predicted data related to option premiums in future periods, such as predicted exercise prices, futures prices, volatility calculated from historical data, and annualized expiration time of futures products in the future period;

[0051] S306, calculating the object reference threshold according to the exercise price, the futures price, the annualized expiration time of the futures product and the maximum fluctuation range in the past three years;

[0052] S308, comparing the predicted option cost with the object reference threshold to detect whether the option cost is higher than the object reference threshold. If it is higher than the object reference threshold, it is considered that the predicted option cost is inaccurate.

[0053] Through the embodiments of the present application, a normal distribution calculation is performed on a transaction parameter set of a target object to obtain an object reference value, and anomaly detection is performed on a second transaction parameter based on the object reference value and a set of transaction parameters to obtain an anomaly detection result. Since the transaction parameter set includes a set of transaction parameters of the target object and the fluctuation range of the first transaction parameter, the anomaly detection result is more accurate, thereby improving the accuracy of anomaly detection, thereby solving the technical problem of low detection accuracy of predicted transaction parameters in related technologies.

[0054] As an optional scheme, before using the object parameters of the target object to generate the trading parameter set of the target object, the method also includes: obtaining the historical parameter values ​​of the first trading parameter in multiple historical time periods; using the historical parameter values ​​in each historical time period to calculate the volatility in each historical time period to obtain multiple first volatilities; determining the average value of the multiple first volatilities as the first fluctuation range, wherein the historical parameter values ​​in each historical time period have the same calculation weight in the process of calculating the first fluctuation range; calculating the second fluctuation range in multiple historical time periods according to the weight corresponding to each historical time period and multiple first volatilities; inputting the historical parameter values ​​in multiple historical time periods into the target statistical model to obtain a third fluctuation range, wherein the target statistical model is a statistical model that analyzes the volatility in time series data; and calculating the target fluctuation range based on the first fluctuation range, the second fluctuation range and the third fluctuation range.

[0055] Optionally, in the embodiment of the present application, the target fluctuation range may be volatility. In the present embodiment, the volatility is calculated using historical price data, and methods such as simple moving average method, exponential moving average method or GARCH model may be used.

[0056] Optionally, the historical time period in this embodiment can be of any length, such as daily, weekly, monthly or annual. In each determined historical time period, relevant parameter values ​​are collected. These parameter values ​​can be stock prices, interest rates, exchange rates, etc., depending on the analysis objectives. By calculating the volatility of each historical time period, multiple first volatility rates are obtained. These volatility rates are a measure of the fluctuation of the parameter values ​​in the time period. These first volatility rates are averaged to obtain a first fluctuation range. In this process, the calculation weight of each first volatility rate is the same, that is, the contribution of each time period to the final fluctuation range is equal. This average value is determined as the first fluctuation range, which represents the average degree of fluctuation of the parameter value in all considered historical time periods. This embodiment provides a comprehensive fluctuation range measurement that can be used to compare the volatility of different time periods.

[0057] Alternatively, the exponential moving average is a time series forecasting method that smoothes and forecasts data by weighted moving average. In the exponential moving average, weights are numerical values ​​used to give different importance to different historical data points. Weights usually decrease over time, i.e., more recent data points have higher weights. The first volatility can be a standard deviation, variance, or other statistic that measures volatility. The second volatility is the calculated volatility expectation for future time periods.

[0058] The calculation steps include:

[0059] S1, determine the weight: select a decay factor (α), usually between 0 and 1. The weight can be calculated by the following formula: \[W_t=\alpha*W_{t-1}+(1-\alpha)\], where \(W_t\) is the weight at the current time point and \(W_{t-1}\) is the weight at the previous time point.

[0060] S2, calculate the first volatility: For each historical time period, calculate its volatility. This can be done by calculating the standard deviation or other statistics.

[0061] S3, apply the exponential moving average method: for the volatility in each historical time period, use the weight and decay factor to calculate the weighted average volatility. This can be done by the following formula: \[EMA_t=\alpha*(V_t-EMA_{t-1})+EMA_{t-1}\], where \(EMA_t\) is the exponential moving average volatility at the current time point, \(V_t\) is the volatility at the current time point, and \(EMA_{t-1}\) is the exponential moving average volatility at the previous time point.

[0062] S4, predicting the second volatility: Based on the calculated exponential moving average volatility, the volatility of the future time period can be predicted. This is usually done by taking the last calculated exponential moving average volatility as the predicted value.

[0063] The advantage of this method is that it can smooth the data and reduce the impact of short-term fluctuations while maintaining sensitivity to the latest data. However, in actual applications, it may be necessary to adjust the attenuation factor and other parameters according to specific circumstances.

[0064] Optionally, the target statistical model includes but is not limited to a GARCH (Generalized AutoregressiveConditional Heteroskedasticity) model, which is a statistical model used to describe the volatility (i.e., risk or uncertainty) of time series data. The basic idea of ​​the GARCH model is to predict future volatility through past information (e.g., historical volatility). It takes into account the autocorrelation of time series data, that is, current volatility may be affected by past volatility.

[0065] The volatility in this embodiment is calculated using historical data, and other methods such as implied volatility may also be used.

[0066] This embodiment can more comprehensively and accurately reflect market fluctuations and provide a reference for investment decisions by comprehensively considering the volatility, weights and target statistical models in multiple historical time periods.

[0067] As an optional scheme, a normal distribution calculation is performed on the trading parameter set to obtain an object reference threshold of the target object within a target prediction time period, including: inputting the set of trading parameters included in the trading parameter set and the target fluctuation range into a normal distribution model to obtain the object reference threshold, wherein the normal distribution model is determined by performing normal distribution modeling on the set of trading parameters and the target fluctuation range.

[0068] Optionally, the normal distribution model calculates the object reference threshold by the following formula: C = K*np.exp(-r*T)*N_d1-F*N_d2; wherein, C is used to represent the object reference threshold, d1 = (np.log(F / K)+(r+0.5*sigma**2)*T) / (sigma*np.sqrt(T)), d2 = d1-sigma*np.sqrt(T), and N_d1 = norm.c df(d1), N_d2=norm.cdf(d2), K, F, r, and T are all parameters in the set of trading parameters, sigma is used to represent the target volatility, np.log is used to represent the logarithm of F / K, np.sqrt is used to represent the square root of T, and norm.cdf is used to represent the cumulative distribution function of the normal distribution, which is used to calculate the probability that the normal distribution random variable is less than or equal to d1 or d2. For example, assuming that the parameters of exercise price K=100, futures price F=105, risk-free interest rate (annualized) r=0.03, expiration time (year) T=0.5, and maximum volatility sigma_max in the past three years=0.25; the above parameter values ​​are substituted into the formula to calculate the object reference threshold. This embodiment can calculate an accurate object reference threshold through the normal distribution model, saving computing resources.

[0069] In an exemplary embodiment, anomaly detection is performed on a second transaction parameter based on the object reference threshold and the set of transaction parameters to obtain a first anomaly detection result, including: in response to a parameter detection request, obtaining the second transaction parameter; inputting the object reference threshold, the set of transaction parameters and the second transaction parameter into a target classification model to obtain a classification result, wherein the target classification model includes a feature extraction module and a classification module, the feature extraction module is used to extract features of the object reference threshold and features of the set of transaction parameters to obtain a first feature description matrix, and is used to extract features of the second transaction parameter to obtain a second feature description matrix, and the first feature description matrix and the second feature description matrix are merged to obtain a target feature description matrix, and the classification module is used to perform a classification operation on the target feature description matrix to obtain a classification label; and the second transaction parameter is detected based on the classification label to obtain the first anomaly detection result.

[0070] Optionally, the second transaction parameter includes but is not limited to transaction volume, transaction frequency, transaction price, transaction fee, etc. For example, in this embodiment, it can be the option premium of an option. The first feature description matrix and the second feature description matrix refer to organizing all the eigenvalues ​​of each object into a matrix, in which each row represents an object and each column represents a feature. For example, if there are three objects and five features, the matrix size is 3x5. The information in the first feature description matrix and the second feature description matrix are both eigenvectors.

[0071] Optionally, when merging the first feature description matrix and the second feature description matrix, the following can be done: Determine the dimension of the merged target feature description matrix. Assume that the first feature description matrix has m rows and n columns, and the second feature description matrix has m rows and p columns. The merged target feature description matrix will have m rows and (n+p) columns. Create an empty matrix with the same number of rows and columns as the merged target feature description matrix. Copy each element of the first feature description matrix to the corresponding position of the target feature description matrix, and then copy the elements of the second feature description matrix to the remaining positions of the target feature description matrix according to the same row and column offsets.

[0072] Here is a specific example:

[0073] S1, assume that there are two feature description matrices:

[0074] The first feature description matrix M1: 1 2 3; 4 5 6; 7 8 9;

[0078] The second feature description matrix M2: 10 11; 12 13; 14 15;

[0082] S2, merge according to the above steps:

[0083] Determine the dimension of the target feature description matrix: 3 rows and 5 columns.

[0084] Initialize the target feature description matrix M_target: 0 0 0 0 0; 0 0 0 0 0; 0 0 0 0 0;

[0088] Merge feature description matrices:

[0089] M1[0][0]->M_target[0][0]->1;

[0090] M1[0][1]->M_target[0][1]->2;

[0091] M1[2][2]->M_target[2][4]->9;

[0092] M2[0][0]->M_target[0][2]->10;

[0093] M2[0][1]->M_target[0][3]->11;

[0094] M2[2][1]->M_target[2][5]->15;

[0095] The final merged target feature description matrix M_target is: 1 2 10 11 0; 4 5 12 13 0;

[0098] 7 8 14 15 0.

[0099] The first feature description matrix and the second feature description matrix are combined to form a comprehensive target feature description matrix.

[0100] In an exemplary embodiment, after detecting the second transaction parameter based on the classification label and obtaining the first anomaly detection result, the method further includes: when it is determined that the second transaction parameter is abnormal based on the classification label, calculating a first difference between the second transaction parameter and the object reference threshold; sending the first difference to the target prediction platform, so that the target prediction platform predicts abnormal prompt information based on the abnormality level, wherein the abnormal prompt information includes abnormal description information, abnormal cause, abnormal impact and abnormal response measures, and the abnormal prompt information is also used to instruct the target prediction platform to adjust the transaction factor of the transaction parameter predicted for the target object.

[0101] Optionally, assume that the second trading parameter predicted by the target prediction platform is the option premium of the option transaction. Assume that the nominal principal of the option ((P)) is 1,000,000 yuan. The object reference threshold is set to 1.5% of the nominal principal, that is, 15,000 yuan. The profit to the customer ((Profit)) needs to be calculated in some way, but for simplicity, we assume that this value is already given. The calculation process includes:

[0102] Object reference threshold = (0.015\times P);

[0103] object_reference_threshold = (0.015\times 1000000);

[0104] Object reference threshold = 15,000 yuan.

[0105] Assuming that the customer is profitable, there are two situations:

[0106] Case 1: The profit from the customer is 10,000 yuan (not exceeding the threshold);

[0107] Case 2: Customer profit is 20,000 yuan (exceeding the threshold) ；

[0108] Comparing the customer profit with the reference threshold, case 1: 10,000 yuan < 15,000 yuan, so the transaction is not an abnormal option pricing transaction.

[0109] Case 2: 20,000 yuan>15,000 yuan, so the transaction is an option transaction and the pricing is an abnormal transaction.

[0110] Optionally, as described above, when the first difference is 5000, the abnormality level determined by the target prediction platform is a high abnormality, and the abnormality prompt information issued includes abnormality description information. The abnormality description information may be determined by the target prediction platform.

[0111] Abnormal description information includes but is not limited to: the pricing of option premiums is high, and there may be illegal operations. The target prediction platform obtains the volatility of the underlying asset and analyzes whether the abnormality is related to the volatility of the underlying asset. The target prediction platform calculates the intrinsic value of the option and analyzes whether the abnormality is related to the profit immediately obtained when the option is exercised. The target prediction platform obtains the option premium, which includes time value, that is, the value of the remaining expiration time of the option. The longer the option is to expire, the higher the time value and the higher the option premium. This is because the possibility of change in the price of the underlying asset increases over time, and the potential value of the option also increases accordingly. Analyze whether the abnormality is related to time value. The target prediction platform determines whether the option premium is affected by the risk-free interest rate. If the risk-free interest rate rises, the option premium may also rise. This is because investors seek higher returns in the financial market, resulting in an increase in demand for options, thereby pushing up option premiums. The supply and demand relationship in the option market will also affect option premiums. If the market demand for a certain type of option increases and the supply is limited, the option premium may rise. The closer the option's strike price is to the market price of the underlying asset, the higher the option premium. This is because options with strike prices close to the market price have higher intrinsic value and investment attractiveness. Transaction fees and commissions charged by exchanges or brokers will also affect option premiums. Economic events may affect market sentiment and investors' risk expectations, thereby affecting option premiums. Countermeasures include but are not limited to: trying to use different option premium pricing models, comparing the results of different models, and selecting the most appropriate model for pricing.

[0112] Abnormal alert information is also used to guide the target prediction platform to adjust the trading parameters in its prediction model. This may include risk assessment, trading strategy, market trends and other factors. The target prediction platform can also monitor and update the mechanism in real time to continuously optimize the prediction model and abnormal alert information according to new data and market changes.

[0113] This embodiment predicts abnormal prompt information based on the abnormal level through the target prediction platform, thereby improving the accuracy of detection.

[0114] In an exemplary embodiment, after performing anomaly detection on the second trading parameter based on the object reference threshold and the set of trading parameters to obtain a first anomaly detection result, the method further includes: updating the multiple historical time periods according to a preset time period to update the target fluctuation range; re-performing normal distribution calculation on the trading parameter set according to the updated target fluctuation range to determine an updated object reference threshold; re-performing anomaly detection on the second trading parameter based on the updated object reference threshold and the set of trading parameters to obtain a second anomaly detection result; comparing the first anomaly detection result with the second anomaly detection result to obtain a comparison result; and determining the target detection result of the second trading parameter based on the comparison result.

[0115] Optionally, the price and market index of the target object in different time periods are different, and the calculated fluctuation range is also different. By appropriately adjusting the historical time period and combining the fluctuation ranges of multiple historical time periods, a more objective fluctuation range can be determined.

[0116] It should be noted that the method may include but is not limited to periodically acquiring historical data in different historical time periods, updating the historical data, and then using the latest acquired historical data to calculate the fluctuation range. For example, historical data may be acquired every 10 days, the fluctuation range may be re-determined, and the object reference threshold may be updated.

[0117] For example, the above historical data is obtained at a preset time node, the historical data is updated, and then the target fluctuation range is calculated using the latest historical data to determine the latest object reference threshold. Then, anomaly detection is performed on the second transaction parameter based on the latest object reference threshold and a set of transaction parameters. For example, historical data is obtained at 9 o'clock every Monday, the target fluctuation range is re-determined, and the object reference threshold is updated.

[0118] Optionally, in an embodiment of the present application, the abnormal points identified by using different object reference thresholds are compared to see whether they are consistent. If both results indicate that the second transaction parameter is abnormal, the probability that the second transaction parameter is an abnormal parameter is relatively high. If the two results are different, these differences need to be further analyzed. Possible reasons include but are not limited to different detection algorithms, different parameter settings, differences in data preprocessing, etc.

[0119] The above is only an example, and this application does not make any specific limitation.

[0120] Obviously, the embodiments described above are only part of the embodiments of the present application, rather than all the embodiments.

[0121] The present application is described in detail below with reference to specific embodiments:

[0122] Exemplarily, the embodiments of the present application can be applied to application scenarios for anomaly detection of option premiums, including but not limited to detecting the impact of different factors that may affect option premiums, so as to provide more accurate and comprehensive pricing of option premiums and help investors make more informed decisions.

[0123] Modeling based on commodity option premiums and taking the maximum volatility of each product in the past three years, calculating the "maximum theoretical value" of option premiums for each transaction, that is, the object reference threshold, can ensure that the calculation of option premiums is closer to the actual situation and reduce risks. Compare and check transactions that exceed the "maximum theoretical value" of option premiums.

[0124] Calculate the "maximum theoretical value" of commodity option premiums based on the Black-Scholes model and compare and check actual transactions that exceed this value, such as Figure 4 Said, comprising the following steps:

[0125] S401, establish the Black-Scholes model, including:

[0126] Data collection: Collect historical price data of various futures products in the past three years.

[0127] Calculate the volatility of each product in the past three years and select the maximum value as the "maximum volatility".

[0128] Parameter setting: Determine the strike price of each product. Determine the current futures price of each product. Set the risk-free rate. You can use the corresponding risk-free rate in the market, such as the government bond rate. Set the expiration time, that is, the remaining expiration time of the option contract.

[0129] S402, Black-Scholes model calculation:

[0130] Use the Black-Scholes model formula and substitute the collected parameters (including "maximum volatility", predicted price parameters, etc.) into the formula to calculate the "maximum theoretical value" of each option variety.

[0131] The specific composition structure of the established model: Input parameters: exercise price, futures price, risk-free interest rate, expiration time, and maximum volatility.

[0132] Calculation process: Calculate d1 and d2. Use the NORMSDIST function (or similar cumulative distribution function) to calculate N(d1) and N(d2). Substitute into the Black-Scholes model formula to calculate the option premium.

[0133] Running process: Write a Python script or program to read historical price data, calculate volatility, and set parameters. Call the Black-Scholes model function and pass in parameters for calculation.

[0134] The calculated "highest theoretical value" is stored for subsequent comparison.

[0135] Algorithms used: Volatility calculation: Volatility is usually calculated using historical price data, and methods such as simple moving average, exponential moving average or GARCH model can be used.

[0136] Black-Scholes model: This is the main algorithm for calculating option premiums. It estimates option prices based on a series of assumptions (such as the underlying asset price follows a log-normal distribution and the market is frictionless).

[0137] S403, comparison check: Collect the option premiums of actual transactions. Compare the actual option premiums with the calculated "maximum theoretical value". Identify and report those transactions where the actual option premium exceeds the "maximum theoretical value".

[0138] This embodiment can improve the efficiency of detection, shorten the detection cycle, and meet the market's demand for real-time performance through automated data processing and logical analysis. Compared with manual detection, automated detection reduces the interference of subjective judgment and improves the accuracy of detection. At the same time, through preset thresholds and algorithms, the compliance of transaction behavior can be judged more objectively.

[0139] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present application can essentially be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, or optical disk), and includes a number of instructions for enabling a front-end device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods of each embodiment of the present application. The information collected in the above embodiments is information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of the relevant data comply with the relevant laws, regulations and standards of the relevant countries and regions, take necessary confidentiality measures, do not violate public order and good customs, and provide corresponding operation entrances for users to choose to authorize or refuse. The above embodiments are based on big data analysis of personal information, and / or use artificial intelligence technology to make automated decisions to make decisions that have a significant impact on personal rights and interests. The technical solution will provide users with corresponding operation entrances for users to choose to agree or refuse the automated decision-making results; if the user chooses to refuse, the expert decision-making process will be entered.

[0140] According to another aspect of the embodiment of the present application, a parameter detection device is also provided, which is applied to a data detection device, and the data detection device is connected to a target prediction platform, and the target prediction platform is used to predict the object parameters of multiple objects based on multiple transaction factors. The device is used to implement the above-mentioned embodiments and preferred embodiments, and the descriptions have been made and will not be repeated. As used below, the term "module" can implement a combination of software and / or hardware of a predetermined function. Although the device described in the following embodiments is preferably implemented in software, the implementation of hardware, or a combination of software and hardware is also possible and conceived.

[0141] Figure 5 is a structural block diagram of a parameter detection device according to an embodiment of the present application, such as Figure 5 As shown, the device comprises:

[0142] A first acquisition module 52 is used to acquire object parameters of a target object in a target prediction time period from the target prediction platform, wherein the target object is any one of the multiple objects;

[0143] A first generating module 54 is configured to generate a trading parameter set of the target object using the object parameters of the target object, wherein the trading parameter set includes a set of trading parameters of the target object and a target fluctuation range of a first trading parameter, and the target fluctuation range is a value of the first trading parameter that is greater than a preset range among continuous fluctuation ranges of the first trading parameter in multiple historical time periods;

[0144] A first calculation module 56, configured to perform normal distribution calculation on the transaction parameter set to obtain an object reference threshold value of the target object within a target prediction time period;

[0145] The first detection module 58 is used to perform anomaly detection on the second transaction parameter based on the object reference threshold and the set of transaction parameters to obtain a first anomaly detection result, wherein the second transaction parameter is a parameter generated by the target prediction platform during the process of performing a risk prediction operation on the target object within the target prediction time period.

[0146] In an exemplary embodiment, the above-mentioned device also includes: a second acquisition module, which is used to obtain the historical parameter values ​​of the first trading parameter in multiple historical time periods before generating the trading parameter set of the target object using the object parameters of the target object; a first calculation module, which is used to calculate the volatility in each historical time period using the historical parameter values ​​in each historical time period to obtain multiple first volatility; a first determination module, which is used to determine the average value of the multiple first volatility as the first volatility, wherein the historical parameter values ​​in each historical time period have the same calculation weight in the process of calculating the first volatility; a second calculation module, which is used to calculate the second volatility in multiple historical time periods according to the weight corresponding to each historical time period and multiple first volatility; a first input module, which is used to input the historical parameter values ​​in multiple historical time periods into a target statistical model to obtain a third volatility, wherein the target statistical model is a statistical model by analyzing the volatility in time series data; a third calculation module, which is used to calculate the target volatility based on the first volatility, the second volatility and the third volatility.

[0147] In an exemplary embodiment, the above-mentioned first calculation module includes: a first input unit, used to input the set of trading parameters included in the trading parameter set and the target fluctuation range into the normal distribution model to obtain the object reference threshold, wherein the normal distribution model is determined by performing normal distribution modeling on the set of trading parameters and the target fluctuation range.

[0148] In an exemplary embodiment, the normal distribution model calculates the object reference threshold by the following formula: C = K*np.exp(-r*T)*N_d1-F*N_d2; wherein, C is used to represent the object reference threshold, d1 = (np.log(F / K)+(r+0.5*sigma**2)*T) / (sigma*np.sqrt(T)), d2 = d1-sigma*np.sqrt(T), N_d1 = no rm.cdf(d1), N_d2=norm.cdf(d2), K, F, r, and T are all parameters in the set of trading parameters, sigma is used to represent the target volatility, np.log is used to represent the logarithm of F / K, np.sqrt is used to represent the square root of T, and norm.cdf is used to represent the cumulative distribution function of the normal distribution, which is used to calculate the probability that a normally distributed random variable is less than or equal to d1 or d2.

[0149] In an exemplary embodiment, the above-mentioned first detection module includes: a first acquisition unit, used to obtain the second transaction parameter in response to a parameter detection request; a second input unit, used to input the object reference threshold, the set of transaction parameters and the second transaction parameter into a target classification model to obtain a classification result, wherein the target classification model includes a feature extraction module and a classification module, the feature extraction module is used to extract features of the object reference threshold and features of the set of transaction parameters to obtain a first feature description matrix, and is used to extract features of the second transaction parameter to obtain a second feature description matrix, and merge the first feature description matrix and the second feature description matrix to obtain a target feature description matrix, and the classification module is used to perform a classification operation on the target feature description matrix to obtain a classification label; the first detection unit is used to detect the second transaction parameter based on the classification label to obtain the first anomaly detection result.

[0150] In an exemplary embodiment, the above-mentioned device also includes: a fourth calculation module, which is used to detect the second transaction parameter based on the classification label, and after obtaining the first anomaly detection result, when it is determined that the second transaction parameter is abnormal based on the classification label, calculate the first difference between the second transaction parameter and the object reference threshold; a first prediction module, which is used to send the first difference to the target prediction platform, so that the target prediction platform predicts abnormal prompt information based on the abnormality level, wherein the abnormal prompt information includes abnormal description information, abnormal causes, abnormal impacts and abnormal countermeasures, and the abnormal prompt information is also used to instruct the target prediction platform to adjust the transaction factor of the transaction parameter predicted for the target object.

[0151] In an exemplary embodiment, the above-mentioned device also includes: a first updating module, which is used to perform anomaly detection on the second trading parameter based on the object reference threshold and the set of trading parameters, and after obtaining the first anomaly detection result, update the multiple historical time periods according to a preset time period to update the target fluctuation range; a first execution module, which is used to re-execute the normal distribution calculation on the trading parameter set according to the updated target fluctuation range to determine the updated object reference threshold; a second detection module, which is used to re-execute anomaly detection on the second trading parameter based on the updated object reference threshold and the set of trading parameters to obtain a second anomaly detection result; a first comparison module, which is used to compare the first anomaly detection result with the second anomaly detection result to obtain a comparison result; and a first determination module, which is used to determine the target detection result of the second trading parameter based on the comparison result.

[0152] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0153] It should be noted that the above modules can be implemented by software or hardware. For the latter, it can be implemented in the following ways, but not limited to: the above modules are all located in the same processor; or the above modules are located in different processors in any combination.

[0154] An embodiment of the present application further provides a computer-readable storage medium, in which a computer program is stored, wherein the computer program is configured to execute the steps of any of the above method embodiments when running.

[0155] In an exemplary embodiment, the computer-readable storage medium may include, but is not limited to, various media that can store computer programs, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk.

[0156] An embodiment of the present application further provides an electronic device, including a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0157] In an exemplary embodiment, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.

[0158] An embodiment of the present application further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the steps in any one of the above method embodiments are implemented.

[0159] An embodiment of the present application further provides another computer program product, including a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in any of the above method embodiments are implemented.

[0160] The embodiments of the present application also provide a computer program, which includes computer instructions, which are stored in a computer-readable storage medium; a processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device performs the steps of any one of the above method embodiments.

[0161] For specific examples in this embodiment, reference may be made to the examples described in the above embodiments and exemplary implementation modes, and this embodiment will not be described in detail herein.

[0162] Obviously, those skilled in the art should understand that the above modules or steps of the present application can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order from that herein, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present application is not limited to any specific combination of hardware and software.

[0163] The above are only preferred embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the principles of the present application shall be included in the protection scope of the present application.

Claims

1. A parameter detection method, characterized in that: Applied to a data detection device, the data detection device is connected to a target prediction platform, the target prediction platform is used to predict object parameters of multiple objects based on multiple transaction factors, the method includes: Acquire object parameters of a target object in a target prediction time period from the target prediction platform, wherein the target object is any one of the multiple objects; Generate a transaction parameter set of the target object using the object parameters of the target object, wherein the transaction parameter set includes a set of transaction parameters of the target object and a target fluctuation range of a first transaction parameter, and the target fluctuation range is a value of the first transaction parameter that is greater than a preset range among continuous fluctuation ranges of the first transaction parameter in multiple historical time periods; Performing normal distribution calculation on the transaction parameter set to obtain an object reference threshold value of the target object within a target prediction time period; Anomaly detection is performed on a second transaction parameter based on the object reference threshold and the set of transaction parameters to obtain a first anomaly detection result, wherein the second transaction parameter is a parameter generated during the process of the target prediction platform performing a risk prediction operation on the target object within the target prediction time period.

2. The method according to claim 1, characterized in that Before generating a transaction parameter set of the target object using the object parameters of the target object, the method further includes: Obtaining historical parameter values ​​of the first transaction parameter in a plurality of the historical time periods; Calculating the volatility in each of the historical time periods using the historical parameter values ​​in each of the historical time periods to obtain a plurality of first volatility; Determine an average value of the plurality of first volatility rates as a first fluctuation range, wherein the historical parameter values ​​in each of the historical time periods have the same calculation weight in the process of calculating the first fluctuation range; Calculating the second fluctuation ranges in the multiple historical time periods according to the weight corresponding to each of the historical time periods and the multiple first fluctuation rates; Inputting historical parameter values ​​in a plurality of the historical time periods into a target statistical model to obtain a third fluctuation range, wherein the target statistical model is a statistical model for analyzing volatility in time series data; The target fluctuation range is calculated based on the first fluctuation range, the second fluctuation range, and the third fluctuation range.

3. The method according to claim 1, characterized in that Performing normal distribution calculation on the transaction parameter set to obtain an object reference threshold of the target object within a target prediction time period includes: The set of trading parameters and the target fluctuation range included in the trading parameter set are input into a normal distribution model to obtain the object reference threshold, wherein the normal distribution model is determined by performing normal distribution modeling on the set of trading parameters and the target fluctuation range.

4. The method according to claim 3, characterized in that The normal distribution model calculates the object reference threshold by the following formula: C=K*np.exp(-r*T)*N_d1-F*N_d2; Among them, C is used to represent the object reference threshold, d1=(np.log(F / K)+(r+0.5*sigma**2)*T) / (sigma*np.sqrt(T)), d2=d1-sigma*np.sqrt(T), N_d1=norm.cdf(d1), N_d2=norm.cdf(d2), K, F, r, and T are all parameters in the set of trading parameters, sigma is used to represent the target fluctuation range, np.log is used to represent the logarithm of F / K, np.sqrt is used to represent the square root of T, and norm.cdf is used to represent the cumulative distribution function of the normal distribution, which is used to calculate the probability that a normally distributed random variable is less than or equal to the d1 or the d2.

5. The method according to claim 1, characterized in that Performing anomaly detection on the second transaction parameter based on the object reference threshold and the set of transaction parameters to obtain a first anomaly detection result includes: In response to the parameter detection request, acquiring the second transaction parameter; Inputting the object reference threshold, the set of transaction parameters, and the second transaction parameter into a target classification model to obtain a classification result, wherein the target classification model includes a feature extraction module and a classification module, the feature extraction module is used to extract features of the object reference threshold and features of the set of transaction parameters to obtain a first feature description matrix, and is used to extract features of the second transaction parameters to obtain a second feature description matrix, and the first feature description matrix and the second feature description matrix are combined to obtain a target feature description matrix, and the classification module is used to perform a classification operation on the target feature description matrix to obtain a classification label; The second transaction parameter is detected based on the classification label to obtain the first anomaly detection result.

6. The method according to claim 5, characterized in that After detecting the second transaction parameter based on the classification label and obtaining the first anomaly detection result, the method further includes: In a case where it is determined based on the classification label that the second transaction parameter is abnormal, calculating a first difference between the second transaction parameter and the object reference threshold; The first difference is sent to the target prediction platform so that the target prediction platform can predict abnormal prompt information based on the abnormal level, wherein the abnormal prompt information includes abnormal description information, abnormal cause, abnormal impact and abnormal response measures, and the abnormal prompt information is also used to instruct the target prediction platform to adjust the transaction factor of the transaction parameters predicted for the target object.

7. The method according to claim 1, characterized in that After performing anomaly detection on the second transaction parameter based on the object reference threshold and the set of transaction parameters to obtain a first anomaly detection result, the method further includes: Updating the plurality of historical time periods according to a preset time period to update the target fluctuation range; Re-execute normal distribution calculation on the transaction parameter set according to the updated target fluctuation range to determine an updated object reference threshold; re-perform anomaly detection on the second transaction parameter based on the updated object reference threshold and the set of transaction parameters to obtain a second anomaly detection result; comparing the first abnormality detection result and the second abnormality detection result to obtain a comparison result; A target detection result of the second transaction parameter is determined based on the comparison result.

8. A parameter detection device, characterized in that: Applied to a data detection device, the data detection device is connected to a target prediction platform, the target prediction platform is used to predict object parameters of multiple objects based on multiple transaction factors, the device comprises: A first acquisition module is used to acquire object parameters of a target object in a target prediction time period from the target prediction platform, wherein the target object is any one of the multiple objects; A first generating module, configured to generate a trading parameter set of the target object using the object parameters of the target object, wherein the trading parameter set includes a set of trading parameters of the target object and a target fluctuation range of a first trading parameter, wherein the target fluctuation range is a value greater than a preset range among continuous fluctuation ranges of the first trading parameter in multiple historical time periods; A first calculation module is used to perform normal distribution calculation on the transaction parameter set to obtain an object reference threshold value of the target object within a target prediction time period; The first detection module is used to perform anomaly detection on a second transaction parameter based on the object reference threshold and the set of transaction parameters to obtain a first anomaly detection result, wherein the second transaction parameter is a parameter generated during the process of the target prediction platform performing a risk prediction operation on the target object within the target prediction time period.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, wherein the computer program implements the steps of the method described in any one of claims 1 to 7 when executed by a processor.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method described in any one of claims 1 to 7 are implemented.

Images