Online car-hailing price elasticity evaluation method, device and equipment and storage medium

Abstract

The invention relates to an online car-hailing price elasticity evaluation method, apparatus and device, and a storage medium. The method comprises the steps of obtaining transaction data of an online car-hailing platform; wherein the transaction data comprises inquiry time, order placing time, a vehicle type, mileage, a pre-estimated price, order completion information and supply and demand information; selecting features based on the transaction data according to a principle that the greater the correlation with price elasticity is, the more priority is; constructing a DML price elastic model; wherein the DML price elastic model comprises two sub-modules, and one sub-module is used for estimating the online car-hailing price serving as a processing variable; the other sub-module is used for estimating a riding demand serving as a response variable; based on the transaction data, a residual error fitting mode is adopted, and a Cross-fitting algorithm is used to train a DML price elasticity model; and based on the trained DML price elasticity model, carrying out price elasticity evaluation on the online car-hailing price to obtain an evaluation result.

Description

Technical Field

[0001] The present invention relates to the technical field of elasticity assessment, and in particular to a method, device, equipment and storage medium for assessing the price elasticity of an online car-hailing service. Background Art

[0002] In the modern business environment, pricing strategies are crucial to a company's profitability and market competitiveness. Price elasticity assessment is a key tool for developing effective pricing strategies. It quantifies the impact of price changes on consumer demand for products or services. Traditional price elasticity assessment methods mainly rely on economic theory and statistical analysis, but the results obtained in this way are a great waste of manpower and are not accurate. Summary of the invention

[0003] In view of this, a method, device, equipment and storage medium for evaluating the price elasticity of an online car-hailing service are provided.

[0004] The present invention adopts the following technical solution: An embodiment of the present invention provides a method for evaluating the price elasticity of an online car-hailing service, the method comprising: Obtain transaction data from online ride-hailing platforms; The transaction data includes inquiry time, order time, vehicle model, mileage, estimated price, order completion information, and supply and demand information; Based on the transaction data, selecting features according to the principle of giving priority to features with greater relevance to price elasticity; Constructing a DML price elasticity model; wherein the DML price elasticity model includes: two submodules, one of which is used to estimate the price of online ride-hailing vehicles as a treatment variable; and the other submodule is used to estimate the ride demand as a response variable; Based on the transaction data, the DML price elasticity model is trained by using a residual fitting method and a Cross-fitting algorithm; Based on the trained DML price elasticity model, the price elasticity of online car-hailing prices is evaluated and the evaluation results are obtained.

[0005] In some embodiments, the method further includes: pre-processing the transaction data to process missing data and abnormal data in the data.

[0006] In some embodiments, the selecting of features based on the transaction data according to the principle of greater priority for features with greater correlation to price elasticity includes: Based on the mutual information algorithm, the transaction data is processed to obtain the correlation between each feature and price elasticity; Based on the correlation between each feature and price elasticity, features are selected according to the principle that the greater the correlation with price elasticity, the higher the priority.

[0007] In some embodiments, it also includes: Discretize continuous features; Perform periodic encoding processing on time features.

[0008] In some embodiments, the residuals used for fitting include: the residuals of online car-hailing prices as a treatment variable and the residuals of ride demand as a response variable.

[0009] In some embodiments, the training of the DML price elasticity model by means of a Cross-fitting algorithm includes: The DML price elasticity model is trained with the help of multiple rounds of Cross-fitting algorithm.

[0010] The embodiment of the present invention further provides a device for evaluating the price elasticity of an online car-hailing service, comprising: The acquisition module is used to obtain transaction data of the online car-hailing platform; The transaction data includes inquiry time, order time, vehicle model, mileage, estimated price, order completion information, and supply and demand information; A selection module, for selecting features based on the transaction data according to the principle of having a higher priority for features with greater relevance to price elasticity; Constructing a module to construct a DML price elasticity model; wherein the DML price elasticity model includes: two sub-modules, one of which is used to estimate the price of online ride-hailing vehicles as a treatment variable; and the other sub-module is used to estimate the ride demand as a response variable; A training module, for training the DML price elasticity model based on the transaction data by adopting a residual fitting method and a Cross-fitting algorithm; The evaluation module is used to evaluate the price elasticity of online car-hailing prices based on the trained DML price elasticity model and obtain the evaluation results.

[0011] An embodiment of the present invention further provides an electronic device, including: A processor, and a memory connected to the processor; The memory is used to store a computer program, and the computer program is at least used for the above-mentioned online car-hailing price elasticity evaluation method; The processor is used to call and execute the computer program in the memory.

[0012] An embodiment of the present invention further provides a storage medium storing a computer program. When the computer program is executed by a processor, each step in the above-mentioned method for evaluating the price elasticity of online car-hailing is implemented.

[0013] An embodiment of the present invention also provides a computer program product, including a computer program, which implements the above-mentioned online car-hailing price elasticity evaluation method when executed by a processor. The present invention adopts the above technical scheme to obtain the transaction data of the online car-hailing platform; wherein the transaction data includes inquiry time, order time, vehicle model, mileage, estimated price, order completion information, and supply and demand information; based on the transaction data, select features according to the principle of greater priority with price elasticity; construct a DML price elasticity model; based on the transaction data, adopt the residual fitting method, with the help of Cross-fitting algorithm, train the DML price elasticity model; based on the trained DML price elasticity model, perform price elasticity evaluation on the online car-hailing price to obtain the evaluation result. In this way, by collecting multi-dimensional data such as inquiry time, order time, vehicle model, mileage, estimated price, order completion information and supply and demand information, the transaction situation of the online car-hailing service can be fully understood, providing a basis for in-depth analysis. Real-time updated transaction data can reflect the latest market trends and help enterprises adjust strategies in time to adapt to market changes. Accurate transaction data is a prerequisite for effective price elasticity analysis and helps to improve the accuracy of model prediction. Prioritizing features with large correlation with price elasticity can improve the prediction efficiency and accuracy of the model and avoid noise interference caused by irrelevant features. Through feature selection, the computing resources required for model training and prediction can be reduced, and the analysis efficiency can be improved. Features with strong correlation can better reflect the impact of price changes on demand, providing strong support for enterprises to formulate pricing strategies. Building a model is a key step in understanding and predicting the impact of price changes on demand, which helps to quantify price elasticity. The DML price elasticity model can predict the changes in demand under different price changes, providing a scientific basis for enterprise decision-making. The model is trained with the help of the Cross-fitting algorithm using residual fitting: residual fitting can reduce the deviation of model prediction and improve the accuracy of prediction. The Cross-fitting algorithm improves the generalization ability of the model through cross-validation, ensuring that the model can perform stably on different data sets. Price elasticity evaluation based on the trained DML price elasticity model: the trained DML price elasticity model can provide accurate evaluation results to help enterprises understand the specific impact of price changes on demand. The evaluation results can be directly applied to the formulation of pricing strategies, helping enterprises optimize prices and improve market competitiveness. Through price elasticity evaluation, enterprises can predict the risks that may be brought about by price changes and take measures to manage risks in advance. In summary, by obtaining detailed transaction data, selecting highly correlated features, building and training a DML price elasticity model, and conducting price elasticity evaluation, companies can more scientifically formulate and adjust pricing strategies for online ride-hailing services, improve market response speed and competitiveness, and ultimately maximize profits. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0015] Figure 1 It is a flow chart of a method for evaluating the price elasticity of online car-hailing provided by an embodiment of the present invention; Figure 2 It is a flow chart of a method for evaluating the price elasticity of online car-hailing provided by an embodiment of the present invention; Figure 3 It is a flow chart of a method for evaluating the price elasticity of online car-hailing provided by an embodiment of the present invention; Figure 4 It is a flow chart of a method for evaluating the price elasticity of online car-hailing provided by an embodiment of the present invention; Figure 5 It is a flow chart of a method for evaluating the price elasticity of online car-hailing provided by an embodiment of the present invention; Figure 6 It is a flow chart of a method for evaluating the price elasticity of online car-hailing provided by an embodiment of the present invention; Figure 7 It is a flow chart of a method for evaluating the price elasticity of online car-hailing provided by an embodiment of the present invention; Figure 8 It is a structural schematic diagram of a price elasticity evaluation device for online car-hailing provided by an embodiment of the present invention; Fig. 9 It is a structural schematic diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0016] To make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be described in detail below. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other implementation methods obtained by ordinary technicians in this field without creative work belong to the scope of protection of the present invention.

[0017] First, the application scenario of the embodiment of the present invention is described. Price elasticity evaluation is an important concept in marketing and economics, which reflects the degree of consumer response to changes in commodity prices. In business practice, accurate price elasticity evaluation is crucial for formulating reasonable pricing strategies. With the development of big data and artificial intelligence technology, data-based price elasticity evaluation methods have gradually become a research hotspot.

[0018] In the online car-hailing industry, price elasticity assessment is particularly important because it directly affects the supply and demand balance and profit maximization of the platform. Traditional price elasticity assessment methods such as linear regression analysis have certain limitations, such as difficulty in handling nonlinear relationships and ignoring other influencing factors. In recent years, machine learning methods have been widely used in the field of price elasticity assessment due to their powerful nonlinear relationship modeling capabilities and flexibility.

[0019] At present, the existing technical solutions that are most similar to the present invention mainly focus on improving the accuracy and flexibility of price elasticity evaluation by using statistics and machine learning techniques based on improving traditional methods. The following are several typical existing technical solutions: 1. Evaluation method based on panel data: By collecting data from multiple time periods and building a panel data model to evaluate price elasticity, it can take into account the impact of time series and improve the accuracy of the evaluation. This method has been discussed a lot in academic literature, such as the article published in the Journal of Marketing Research.

[0020] 2. Evaluation method based on machine learning: Use machine learning algorithms (such as random forests, neural networks, etc.) to model the nonlinear relationship between price and sales. It can more accurately capture complex relationships and improve the flexibility of evaluation. This method has many applications in the industry, for example, it is introduced in the book "Machine Learning in Business Analytics".

[0021] Limitations of the existing technical solutions Although the above existing technical solutions have improved the limitations of the traditional methods to a certain extent, the following problems still exist: 1. Insufficient comprehensive consideration of multiple factors: Existing technical solutions often find it difficult to comprehensively consider the various factors that affect passenger demand, such as weather conditions, traffic congestion, supply and demand conditions, etc.

[0022] 2. Limited ability to simulate dynamic processes: Existing technical solutions have limited ability to simulate the dynamic changes in the impact of price changes on passenger demand.

[0023] 3. Low real-time performance and flexibility: Existing technical solutions usually require a large amount of historical data to obtain reliable evaluation results. In a rapidly changing market environment, this method has poor real-time performance and flexibility.

[0024] In view of this, the present invention provides a technical solution, proposing a Double Machine Learning method based on online car-hailing transaction observation data to perform price elasticity evaluation, and improving the accuracy, flexibility and real-time performance of price elasticity evaluation by introducing more advanced machine learning algorithms and technical means.

[0025] Example Figure 1 A flowchart of a method for evaluating the price elasticity of an online car-hailing vehicle provided by an embodiment of the present invention. The method can be executed by the device for evaluating the price elasticity of an online car-hailing vehicle provided by an embodiment of the present invention. Figure 1 , the method may specifically include the following steps: S101. Obtaining transaction data of the online car-hailing platform; The transaction data includes inquiry time, order time, vehicle model, mileage, estimated price, order completion information, and supply and demand information; Specifically, in this step, transaction data is collected from online ride-hailing platforms, which is the basis for price elasticity assessment. Specifically, it includes: Inquiry time: the time when the user initiates the car request.

[0026] Order time: the time when the user actually places the order.

[0027] Car model: The car model selected by the user may affect the price and demand.

[0028] Mileage: The total distance of a trip, which directly relates to the cost.

[0029] Estimated price: The estimated price of the trip given by the platform based on the algorithm.

[0030] Completion information: including trip completion status, user reviews, etc.

[0031] Supply and demand information: The supply and demand conditions at a specific time and place, which affect prices and demand.

[0032] These data provide rich information for analyzing the impact of price changes on demand.

[0033] S102, based on the transaction data, selecting features according to the principle of giving priority to features with greater relevance to price elasticity; In this step, the most relevant features for price elasticity assessment are selected from the collected transaction data. This process includes: Correlation analysis: Use statistical methods (e.g., Pearson correlation coefficient, Spearman rank correlation coefficient) to assess the correlation of each feature with price elasticity.

[0034] Feature selection: Select the features that are most correlated with price elasticity and are most likely to affect changes in quantity demanded.

[0035] Data preprocessing: Perform necessary preprocessing on the selected features, such as normalization, encoding, etc., to improve the performance of the model.

[0036] S103, constructing a DML price elasticity model; wherein the DML price elasticity model includes: two submodules, one of which is used to estimate the price of online car-hailing vehicles as a processing variable; and the other submodule is used to estimate the ride demand as a response variable; In this step, a DML model is constructed which consists of two submodules: Submodule 1: used to estimate the price of online ride-hailing as the treatment variable. Submodule 2: used to estimate the demand for rides as the response variable. These two submodules will be trained in parallel to capture the complex relationship between price and demand.

[0037] S104, based on the transaction data, using a residual fitting method and a Cross-fitting algorithm, training the DML price elasticity model; In this step, the Cross-fitting algorithm is used to train the DML model: Specifically, residual fitting: First, train the two submodules separately to obtain the predicted values ​​of price and demand. Then, calculate the residual, which is the difference between the actual value and the predicted value. Cross-fitting: Divide the dataset into K subsets, take turns using one of the subsets as the test set and the rest as the training set, and repeat K times. This method can reduce the risk of overfitting the model and improve the generalization ability of the model. Model optimization: In each iteration, adjust the model parameters according to the residual of the test set to minimize the prediction error.

[0038] S105. Based on the trained DML price elasticity model, the price elasticity of the online car-hailing price is evaluated to obtain the evaluation result.

[0039] In this step, the trained DML model is used to evaluate price elasticity: Price elasticity evaluation: Input new or existing price data, and the model will output the corresponding demand forecast. Result analysis: Analyze the model output to determine the specific impact of price changes on demand, that is, price elasticity. Decision support: Based on the evaluation results, provide pricing strategy recommendations to online ride-hailing platforms, such as price adjustments, promotional activities, etc.

[0040] Through this series of steps, an accurate and robust price elasticity evaluation model can be obtained, providing a scientific basis for the pricing strategy of online ride-hailing platforms.

[0041] Specifically, refer to Figure 2 The data-based price elasticity evaluation method provided by the present invention mainly introduces the DML method to construct a more accurate price elasticity evaluation model, which solves the problems existing in the traditional method, improves the prediction accuracy and reliability, and is suitable for price strategy formulation in complex market environments such as online car-hailing.

[0042] In the solution provided by this application, platform transaction data is obtained through offline data, including inquiry time, order time, vehicle model, mileage, estimated price, order completion information, supply and demand information, etc. The data collection process is as follows: Figure 3 shown.

[0043] In some embodiments, the process further includes: preprocessing the transaction data to process missing data and abnormal data in the data. That is, for the collected data, records with missing values ​​are deleted and abnormal data are removed. The data preprocessing process is as follows: Figure 4 shown.

[0044] Missing data can be handled as follows: Identify missing data and check missing values: First, use data exploratory analysis to identify which variables contain missing values ​​and the distribution of missing values. Analyze the reasons for missing values: Understanding the reasons for missing data, such as data collection errors, users not providing information, etc., will help decide how to handle these missing values.

[0045] Ways to handle missing values ​​include: Deletion: If the proportion of missing values ​​for a variable is very high, or the missing values ​​are not random, consider deleting the rows or columns with missing values. Filling: For most cases, the following methods can be used to fill missing values: Mean / Median Filling: Replace missing values ​​with the mean or median of the variable, which is suitable for numerical data. Mode Filling: Replace missing values ​​with the mode of the variable, which is suitable for categorical data. Prediction Model Filling: Use machine learning models such as regression and decision trees to predict missing values. Interpolation: For time series data, interpolation can be used to fill missing values.

[0046] Abnormal data processing is as follows: Identifying abnormal data specifically includes: Statistical analysis: Use descriptive statistical analysis (such as box plots, Z scores, etc.) to identify outliers. Visualization methods: Visually identify outliers through visualization methods such as scatter plots and histograms. Processing outliers includes: Deletion: If the outliers are caused by errors and the number is small, you can delete these outliers directly. Correction: If the outliers are caused by data entry errors, you can try to correct these errors. Transformation: Transform the data, such as logarithmic transformation, square root transformation, etc., to reduce the impact of outliers. Retention: In some cases, outliers may represent important information, such as abnormal transactions in fraud detection. In this case, these values ​​should be retained.

[0047] In some embodiments, the feature is selected based on the transaction data according to the principle that the greater the correlation with price elasticity, the higher the priority, including: processing the transaction data based on a mutual information algorithm to obtain the correlation between each feature and price elasticity; based on the correlation between each feature and price elasticity, selecting the feature according to the principle that the greater the correlation with price elasticity, the higher the priority.

[0048] It should be noted that in order to improve the model effect and training speed, feature selection is particularly important, especially selecting unbiased features. Therefore, the collected information is used to calculate the mutual information. In addition, the timestamp is selected as a feature and the timestamp feature is periodically encoded so that the model can better capture the impact of time periodic changes. Feature engineering such as Figure 5 shown.

[0049] Specifically, in the embodiment, using the mutual information algorithm to select the features most relevant to price elasticity is an effective method to improve the prediction accuracy and explanatory power of the price elasticity model. The following are the detailed steps of the process: Collect data: Ensure that you have all relevant transaction data, including but not limited to inquiry time, order time, vehicle model, mileage, estimated price, order completion information, and supply and demand information.

[0050] Data preprocessing: Clean the data, including handling missing values, outliers, normalizing numerical features, and encoding categorical features.

[0051] Calculating the mutual information between features and price elasticity involves: Defining variables: Determine the response variable (price elasticity) and the predictor variable (feature in the transaction data). Applying the mutual information algorithm: Use the mutual information algorithm to calculate the mutual information value between each feature and price elasticity. The mutual information value quantifies the mutual dependence between two variables, and the larger the value, the stronger the correlation.

[0052] Sorting features: Sort all features according to the calculated mutual information values, giving priority to features with high mutual information values.

[0053] Select features: Select the top-ranked features based on model complexity, computing resources, and business needs. Typically, features with higher mutual information values ​​have a greater impact on price elasticity and are therefore more likely to contain information useful for predicting price elasticity.

[0054] Model training: Train a preliminary price elasticity model using the selected features.

[0055] Model evaluation: Evaluate the performance of the model, such as accuracy, stability, and generalization ability. If necessary, you can iteratively adjust feature selection and try adding or removing features to optimize model performance.

[0056] Comprehensive evaluation: Comprehensively consider model performance and business needs to determine the final feature set.

[0057] Documentation: Record the process and results of feature selection, including the mutual information value of each feature and the reasons for the final selection.

[0058] In some embodiments, it also includes: discretizing continuous features; and periodically encoding time features.

[0059] Specifically, discretization is the process of converting continuous variables into categorical variables. This method can reduce the complexity of the data, simplify the model, and sometimes improve the interpretability of the model.

[0060] Periodic encoding is a method of processing periodic features (such as hours, weeks, and months) in time data. It captures the periodic laws of time by converting these features into periodic signals.

[0061] In some embodiments, the residuals used for fitting include: the residuals of the online car-hailing price as a processing variable and the residuals of the ride demand as a response variable. The training of the DML price elasticity model by means of the Cross-fitting algorithm includes: training the DML price elasticity model by means of multiple rounds of the Cross-fitting algorithm.

[0062] The price elasticity model uses a two-stage process to eliminate bias. In the first stage, an arbitrary machine learning model is used to fit whether the transaction is completed ( Y ) and the price coefficient ( T ), respectively, to obtain the predicted values ​​of ( Y ) and ( T ) and calculate the residuals; in the first stage, an arbitrary machine learning model is used to fit the residuals of ( Y ) and ( T ). The Double Machine Learning (DML) elasticity model construction process refers to the following description.

[0063] Model training, the data set is split into training data and validation set to train the model, and 5-fold cross validation is used to evaluate the performance of the model to obtain the causal effect model of price changes on passenger demand. The model training process is as follows Figure 6 shown.

[0064] According to the trained price elasticity model, the price coefficient is estimated based on the input features. The price elasticity prediction process is as follows Figure 7 shown.

[0065] The predicted price elasticity coefficient uses an ordinal relationship to calculate the difference between the actual situation and the estimate, and calculates the area under the elasticity curve (Area Under the Uplift Curve, AUUC) to evaluate the accuracy of price elasticity. The price elasticity evaluation process is described below.

[0066] Specifically, the price elasticity model training process is as follows: 1. Residual fitting: Use any machine learning model to fit the response variable ( Y ) and treatment variable ( T ) separately.

[0067] Get the predicted values ​​and of (Y) and (T), and then calculate the residual sum.

[0068] 2. Fitting residuals: Use any machine learning model to fit the relationship between the residuals and .

[0069] 3.Cross-fitting: Cross-fitting is an important technique to reduce estimation bias.

[0070] Divide the total sample into two subsets ( S_1 ) and ( S_2 ).

[0071] Use (S_1) to estimate and get the residuals, and use (S_2) to estimate and get the residuals.

[0072] Then the above process is repeated by exchanging the roles of the two subsets.

[0073] The final estimate is the average of the two estimates.

[0074] 4. Multi-round Cross-fitting: To further improve the robustness of the estimation, multiple rounds of Cross-fitting can be used, that is, dividing the sample into more subsets, for example, into (K) parts, and then repeating the above process (K) times.

[0075] Specifically, the price elasticity evaluation process includes: Input: In the test set, the price coefficient of 0.95 is the experimental group (W=1), and the price coefficient of 1.0 is the blank group (W=0); whether the order is completed is the response signal Y, 1 for completed orders and 0 for uncompleted orders, and scorei is the price elasticity estimate of a single inquiry. UpliftCurve is an ordinal indicator, and we calculate this ranking result based on the ranking of score.

[0076] calculate: 1. Sort the test set by price elasticity value from large to small, that is, the ones with higher scores are in the front. Based on the sorting results, we can calculate the sum of the responses of the people in the experimental group among the top k, as shown in the following formula:

[0077] Where Y(W) is an indicator function, which is 1 if the order is completed, otherwise 0. Similarly, we can calculate the sum of the responses of the first k people who belong to the blank group:

[0078] Similarly, we can calculate the number of people in the experimental group and the blank group among the top k, represented by NT and NC respectively:

[0079] After getting these four values, we can calculate the uplift value of the kth item. Its calculation formula is as follows:

[0080] The meaning of the kth Uplift value is: the average value of the experimental group among the first k individuals minus the average value of the blank group among the first k individuals. By analogy, we can get the 1st to nth Uplift values, and we can draw a curve based on them.

[0081] Once we have drawn the curve, we can calculate the area under the curve as follows:

[0082] We can normalize this area. The denominator of the normalization is, and the normalized result is

[0083] Figure 8 is a schematic diagram of a structure of a price elasticity evaluation device for online car-hailing provided by another embodiment of the present invention, and the device is suitable for executing a price elasticity evaluation method for online car-hailing provided by an embodiment of the present invention. Figure 8 As shown, the device may specifically include: An acquisition module 81 is used to acquire transaction data of the online car-hailing platform; The transaction data includes inquiry time, order time, vehicle model, mileage, estimated price, order completion information, and supply and demand information; A selection module 82, configured to select features based on the transaction data according to the principle of having a higher priority for features with greater relevance to price elasticity; Constructing module 83, constructing a DML price elasticity model; wherein the DML price elasticity model includes: two submodules, one of which is used to estimate the price of online ride-hailing vehicles as a processing variable; and the other submodule is used to estimate the ride demand as a response variable; A training module 84 is used to train the DML price elasticity model based on the transaction data by adopting a residual fitting method and a Cross-fitting algorithm; The evaluation module 85 is used to evaluate the price elasticity of the online car-hailing price based on the trained DML price elasticity model to obtain an evaluation result.

[0084] The online car-hailing price elasticity evaluation device provided in the embodiment of the present invention can execute the online car-hailing price elasticity evaluation method provided in any embodiment of the present invention, and has the functional modules and beneficial effects corresponding to the execution method. Since it has been elaborated in detail in the method embodiment, it will not be further described in the device embodiment. For details, please refer to the description in the method embodiment.

[0085] The present invention also provides an electronic device. Fig. 9 , Fig. 9 is a schematic diagram of the structure of an electronic device, such as Fig. 9 As shown, the electronic device includes: a processor 91, and a memory 92 connected to the processor 91; the memory 92 is used to store a computer program, and the computer program is at least used to execute the online car-hailing price elasticity evaluation method in an embodiment of the present invention; the processor 91 is used to call and execute the computer program in the memory.

[0086] An embodiment of the present invention further provides a storage medium storing a computer program. When the computer program is executed by a processor, each step of the method for evaluating the price elasticity of an online car-hailing vehicle in the embodiment of the present invention is implemented.

[0087] In addition to the above-mentioned methods and devices, an embodiment of the present application may also be a computer program product, which includes computer program instructions, which, when executed by a processor, enable the processor to execute the steps of the online car-hailing price elasticity evaluation method according to various embodiments of the present application described in the above "Exemplary Method" section of this specification.

[0088] The computer program product may be written in any combination of one or more programming languages ​​to write program codes for performing the operations of the embodiments of the present application, including object-oriented programming languages, such as Java, C++, etc., and conventional procedural programming languages, such as "C" language or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as an independent software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0089] In addition, an embodiment of the present application may also be a computer-readable storage medium on which computer program instructions are stored. When the computer program instructions are executed by a processor, the processor executes the steps of the online car-hailing price elasticity evaluation method according to various embodiments of the present application described in the above "Exemplary Method" section of this specification.

[0090] The computer readable storage medium can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can include, for example, but is not limited to, a system, device or device of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination of the above. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0091] It can be understood that the same or similar parts of the above embodiments can be referenced to each other, and the contents not described in detail in some embodiments can refer to the same or similar contents in other embodiments.

[0092] It should be noted that, in the description of the present invention, the terms "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance. In addition, in the description of the present invention, unless otherwise specified, the meaning of "plurality" refers to at least two.

[0093] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code that includes one or more executable instructions for implementing the steps of a specific logical function or process, and the scope of the preferred embodiments of the present invention includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present invention belong.

[0094] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0095] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0096] In addition, each functional unit in each embodiment of the present invention may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0097] The storage medium mentioned above can be a read-only memory, a magnetic disk or an optical disk, etc.

[0098] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0099] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

Claims

1. A method for evaluating the price elasticity of online car-hailing, characterized in that: include: Obtain transaction data from online ride-hailing platforms; The transaction data includes inquiry time, order time, vehicle model, mileage, estimated price, order completion information, and supply and demand information; Based on the transaction data, selecting features according to the principle of giving priority to features with greater relevance to price elasticity; Constructing a DML price elasticity model; wherein the DML price elasticity model includes: two submodules, one of which is used to estimate the price of online ride-hailing vehicles as a treatment variable; and the other submodule is used to estimate the ride demand as a response variable; Based on the transaction data, the DML price elasticity model is trained by using a residual fitting method and a Cross-fitting algorithm; Based on the trained DML price elasticity model, the price elasticity of online car-hailing prices is evaluated and the evaluation results are obtained.

2. The method according to claim 1, characterized in that Also includes: The transaction data is preprocessed to process missing data and abnormal data in the data.

3. The method according to claim 1, characterized in that The feature selected based on the transaction data according to the principle of greater priority for the feature with greater correlation with price elasticity includes: Based on the mutual information algorithm, the transaction data is processed to obtain the correlation between each feature and price elasticity; Based on the correlation between each feature and price elasticity, features are selected according to the principle that the greater the correlation with price elasticity, the higher the priority.

4. The method according to claim 1, characterized in that: Also includes: Discretize continuous features; Perform periodic encoding processing on time features.

5. The method according to claim 1, characterized in that The residuals used for fitting include: the residual of ride-hailing prices as the treatment variable and the residual of ride demand as the response variable.

6. The method according to claim 1, characterized in that The DML price elasticity model is trained by using the Cross-fitting algorithm, including: The DML price elasticity model is trained with the help of multiple rounds of Cross-fitting algorithm.

7. A price elasticity evaluation device for online car-hailing, characterized in that: include: The acquisition module is used to obtain transaction data of the online car-hailing platform; The transaction data includes inquiry time, order time, vehicle model, mileage, estimated price, order completion information, and supply and demand information; A selection module, for selecting features based on the transaction data according to the principle of having a higher priority for features with greater relevance to price elasticity; Constructing a module to construct a DML price elasticity model; wherein the DML price elasticity model includes: two sub-modules, one of which is used to estimate the price of online ride-hailing vehicles as a treatment variable; and the other sub-module is used to estimate the ride demand as a response variable; A training module, for training the DML price elasticity model based on the transaction data by adopting a residual fitting method and a Cross-fitting algorithm; The evaluation module is used to evaluate the price elasticity of online car-hailing prices based on the trained DML price elasticity model and obtain the evaluation results.

8. An electronic device, characterized in that: include: A processor, and a memory connected to the processor; The memory is used to store a computer program, and the computer program is used at least to execute the online car-hailing price elasticity evaluation method according to any one of claims 1 to 6; The processor is used to call and execute the computer program in the memory.

9. A storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by the processor, each step of the online car-hailing price elasticity evaluation method as described in any one of claims 1-6 is implemented.

10. A computer program product, comprising a computer program, which, when executed by a processor, implements the online car-hailing price elasticity evaluation method described in any one of claims 1 to 6.

Images