WO2022063118A1

WO2022063118A1 - Method and apparatus for determining price sensitivity

Info

Publication number: WO2022063118A1
Application number: PCT/CN2021/119586
Authority: WO
Inventors: 张晓松; 王启凡
Original assignee: 胜斗士(上海)科技技术发展有限公司
Priority date: 2020-09-23
Filing date: 2021-09-22
Publication date: 2022-03-31
Also published as: CN112132617A

Abstract

A method and apparatus for determining price sensitivity. The method comprises: generating a historical data sequence on the basis of historical sales data of a product; removing, from the historical data sequence, a part thereof that is affected by a cyclical factor, so as to generate a processed historical data sequence; on the basis of a plurality of trend polynomials corresponding to a predetermined sales trend and the processed historical data sequence and by using a regression model, determining an optimal trend polynomial corresponding to the sales trend of the product; and determining price sensitivity on the basis of the optimal trend polynomial. By means of the method and the apparatus of the present application, the influence of external factors in a historical data sequence is removed, and the determined price sensitivity between a price entity and a sales volume entity is as close to pure price sensitivity as possible.

Description

Method and apparatus for determining price sensitivity

This application claims the priority of Chinese Patent Application No. 202011008732.7 filed on September 23, 2020. The contents disclosed in the above Chinese patent application are hereby cited in their entirety as a part of this application.

technical field

The present application relates to quantitative evaluation of data, and in particular, to a method and apparatus for determining price sensitivity based on historical sales data for a product.

Background technique

In service industries such as chain restaurants and retail, it is necessary to quantitatively evaluate the price sensitivity between each store's sales data and product selling price to guide the store's overall product price adjustment strategy.

At present, the quantitative analysis of the price sensitivity of stores is mainly based on the traditional economic price sensitivity model. The main variables of the economics price sensitivity model are usually only the product sales volume of the store in a period of time and the product sales price in the period of time. The specific value of the price sensitivity is obtained by calculating the ratio of the change in the sales volume of the product to the change in the sales price of the product (eg, the change in percentage). Normally this value is negative. Generally speaking, stores whose absolute value of price sensitivity value is greater than 1 have high price sensitivity and are not suitable for implementing price increase strategy; stores whose absolute value of price sensitivity value is less than 1 have low price sensitivity and are suitable for implementing price increase strategy ; and stores whose absolute value is equal to 1 are considered to have balanced price sensitivity, thus maintaining the status quo.

However, traditional economics price sensitivity models are mainly based on the assumption that sales volume is only affected by pricing. In a market environment dominated by offline transactions such as the catering retail industry, product sales price is only one of many factors that affect product sales. Other influencing factors include but are not limited to weather factors, seasonal factors, store promotions and discounts, store activities of competitors near the store (such as competitors' new store openings, marketing strategies), and short-term changes in stores (such as temporary store openings). Internal and external decoration construction), long-term changes in the business environment where the store is located, specific events related to the store (such as the holding of concerts, exhibitions, etc.), etc. Under the influence of these complex realistic factors, the relationship between product sales price and sales volume represented by traditional economic price sensitivity models is often inaccurate and unstable, resulting in inaccurate price sensitivity assessment results.

Therefore, there is a need for improved assessment schemes for price sensitivity.

It should be noted that the information disclosed in the above background section is only for enhancing understanding of the background of the application, and therefore may include information that does not form the prior art known to a person of ordinary skill in the art.

SUMMARY OF THE INVENTION

In order to solve at least one disadvantage of the solution of using the above-mentioned traditional economic price sensitivity model for price sensitivity assessment, the present application proposes an improved method and apparatus for determining price sensitivity to obtain a more accurate price sensitivity assessment result. Users provide accurate sales trends to help develop pricing strategies.

According to an aspect of the present application, a method for determining price sensitivity is proposed, comprising:

Generate a historical data sequence based on the historical sales data of the product, the historical data sequence is composed of historical data associated with the historical time in the historical time series, wherein the historical data includes historical sales volume data and historical price data of the product;

Generate a processed historical data series by removing the part affected by periodic factors in the historical data series;

Based on a plurality of trend polynomials corresponding to predetermined sales trends and the processed historical data series, a regression model is used to determine an optimal trend polynomial corresponding to a product's sales trend, and a price sensitivity is determined based on the optimal trend polynomial, wherein the product's The sales volume data is represented as a trend polynomial with price data as the independent variable.

According to another aspect of the present application, an apparatus for determining price sensitivity is also proposed, comprising:

The historical data sequence generating unit is configured to generate a historical data sequence based on the historical sales data of the product, the historical data sequence is composed of historical data associated with the historical time in the historical time series, wherein the historical data includes the historical sales volume data of the product and historical price data;

a historical data sequence processing unit, configured to generate a processed historical data sequence by removing the part affected by the periodic factor in the historical data sequence;

a price sensitivity determination unit configured to use a regression model to determine an optimal trend polynomial corresponding to a sales trend of a product based on a plurality of trend polynomials corresponding to predetermined sales trends and the processed historical data series, and based on the optimal trend The polynomial determines price sensitivity, where the sales volume data for a product is represented as a trend polynomial with price data as the independent variable.

According to yet another aspect of the present application, a computer-readable storage medium is provided, on which a computer program is stored, the computer program comprising executable instructions, which, when executed by a processor, implement the method according to the above .

According to yet another aspect of the present application, an electronic device is provided, comprising a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to execute the executable instructions to implement the above-mentioned method.

It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not limiting of the present application.

Description of drawings

The above and other features and advantages of the present application will become more apparent by describing in detail exemplary embodiments thereof with reference to the accompanying drawings.

1 is a schematic logic flow diagram of a process of determining price sensitivity and formulating a corresponding price strategy according to an example embodiment of the present application;

FIG. 2 is a schematic flowchart of a method for determining price sensitivity according to an example embodiment of the present application;

3 is a schematic structural block diagram of an apparatus for determining price sensitivity according to an exemplary embodiment of the present application; and

FIG. 4 is a schematic block diagram of an electronic device for determining price sensitivity according to an example embodiment of the present application.

detailed description

Example embodiments will now be described more fully with reference to the accompanying drawings. Exemplary embodiments, however, can be embodied in various forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that this application will be thorough and complete, and will fully convey the concept of exemplary embodiments conveyed to those skilled in the art. In the drawings, the size of most elements may be exaggerated or deformed for clarity. The same reference numerals in the drawings denote the same or similar structures, and thus their detailed descriptions will be omitted.

Furthermore, the described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided in order to give a thorough understanding of the embodiments of the present application. However, one skilled in the art will appreciate that the technical solutions of the present application may be practiced without one or more of the specific details, or other methods, elements, etc. may be employed. In other instances, well-known structures, methods, or operations are not shown or described in detail to avoid obscuring aspects of the application.

In this application, the concepts of sales volume entity and price entity are introduced. The sales volume entity is sales volume data representing product sales volume information of sales units during a certain time interval. The sales volume data generally refers to the total product sales volume of the sales unit, and can be obtained by calculating, for example, the sum of the sales volume of each product, weighting the sales volume of different products based on a preset weight, and/or other methods. The price entity is price data representing price information of the product sold in the sales unit during the time interval corresponding to the sales volume entity. In the traditional economic price sensitivity model, the price entity is generally the average of the official sales price or obtained by weighting the official sales price based on a preset weight. According to an embodiment of the present application, the price entity may be the actual selling price of the product (since the customer's sensitivity to the actual selling price is more related to the actual value paid for purchasing the product) or the official selling price. The price entity may also be obtained by processing, such as weighted or weighted average, of the product's sales price based on the product's category and/or sales volume.

Time intervals can be days, weeks, months, quarters, years, hours, minutes, etc., depending on the product category and/or assessment needs. In the following the application is presented with the example of a week, but the application is not limited to the weekly price sensitivity. A sales unit represents an object for evaluating price sensitivity, and can be a sales store located in the same, adjacent, or different sales location, or the sales location itself. In the embodiments of the present application, the method flow of determining the price sensitivity and formulating the corresponding price strategy is described by taking a sales store as an example.

Similar to the traditional economic model, price sensitivity is defined as the ratio of the changes of the sales volume entity and the price entity, that is, the ratio of the changes of the sales volume entity as the sales volume data to the price entity as the price data. . In practice, the sales volume entity is not only related to the sales price, but also affected by other external factors. Although the traditional economic model may take into account that the sales volume entity is affected by factors other than the sales price, because its model is based on the assumption that only the price entity is considered, it cannot exclude other external factors that affect the relationship between the sales volume entity and the price entity. The interference of the interrelationships causes inaccurate and unstable assessment of price sensitivity.

The method for evaluating price sensitivity according to an embodiment of the present application will now be introduced in conjunction with the process of determining price sensitivity and formulating a corresponding price strategy shown in FIG. 1 .

First, at step 110, historical sales data of the store is obtained. The historical sales data of the store is counted at a selected time interval (for example, every week), for example, for the 1st week, the 2nd week, ..., the nth week, the weekly sales volume entity and the price entity of the store are created respectively. The sales volume entity and the price entity are associated with the corresponding week numbers. Here, a week or a week is a time interval between adjacent historical times T, or is referred to as a time unit for measuring the historical time T. If the serial number of the week is denoted as k, the historical time of the kth week can be denoted as T _k . The historical time T _k constitutes a historical time series {T _k } (1≤k≤n, k is a positive integer). To eliminate cyclical factors, especially seasonal factors, on the determination of price sensitivity, at least two years of historical sales data for stores are usually used. Historical sales data can be refined to the category of the product, the actual sale price of the product, and the actual transaction data detail of the date of sale (eg, the week in which the sale was made).

In

steps

121 and 131, respectively, the weekly sales volume entity and the price entity of the store are created. As mentioned above, the Weekly Sales entity can be created by summing the product sales for that week. Weekly Price attributes can be created by taking the average price of a product for the week weighted proportionally to the sales volume of the category. According to the embodiments of the present application, for products with a longer time interval (such as a full month or a year), the weight of the sales volume ratio of the category can also be used as the product sales volume ratio of the historical time _Tk (such as a week). The weights are weighted average to calculate the price entity. The historical data D _k corresponding to the historical time T _k includes the historical sales volume entity (historical sales volume data) M _k and historical price entity (historical price data) P _k of the product in the kth week, then the historical sales data can be expressed as historical Data sequence {D _k } (1≤k≤n, k is a positive integer), wherein historical data D _k =(M _k , P _k ). The historical data D _k is a two-dimensional variable, and its sub-variables M _k and P _k can be one-dimensional or multi-dimensional variables according to the creation method of the corresponding historical sales volume entity and historical price entity. A historical data series {D _k } is associated with a historical time series {T _k }, where D _k is associated with T _k . In this way, the problem of determining price sensitivity based on the historical sales data of a product can be transformed into calculating the sales volume entity (sales data) and the price entity (price based on the historical data sequence {D _k } corresponding to the historical time series {T _k }). data) to guide future price strategies. Viewed another way, it can also be seen as calculating price sensitivity based on a data sequence { _{Dk, Tk} or {Mk} _, _Pk _, _Tk }.

The historical data in the historical data series includes the part affected by periodic factors, and this part of the interference needs to be removed during the price sensitivity calculation process of this application. Periodic factors include holiday factors and seasonal factors. The following will remove the holiday factors and seasonal factors respectively to obtain the processed historical data series {D _k ^* }.

In step 122, the sales volume entities affected by holiday factors are removed from the historical data sequence {D _k }, that is, the historical data {M _i , P _i } at the historical time T _i are removed, where 1≤i≤n, i is a positive integer. The holiday factors include Gregorian holidays that appear on a fixed date each year and lunar holidays that do not appear on a fixed date. Since the dates of the holidays on the lunar calendar are not fixed, it is inconvenient to remove them in the periodic time series processing method based on statistics, so this part of the historical data elements D _i need to be removed from the historical data sequence {D _k } separately. In the selection of the removed historical data, the historical data D _i corresponding to all the lunar holidays and/or the solar holidays can be removed, or the lunar holidays and the solar holidays related to the year can be cross-referenced, which will only result in weekly The historical data D _i corresponding to the holidays of the lunar calendar and/or the solar calendar with the misplaced holidays are removed. By cleaning the historical data of the corresponding weeks, it can be ensured that all historical data elements in the processed historical data sequence {D _k ^* } formed by the remaining historical data can be equivalent in terms of holiday components. After removing the elements of the historical data sequence {D _k ^* } affected by holiday factors, the number of elements is generally smaller than the number of elements of the original historical data sequence {D _k }, and the number of elements of the corresponding historical time series {T _k ^* } is also smaller than that of the original historical data sequence {D k * }. The number of elements of the historical time series {T _k }. In other words, the number of data items in the series is reduced after the holiday factor data processing.

In step 122, components affected by seasonal factors in the value of the historical sales volume entity _Mk in the historical data _Dk are removed. Seasonal factors include, for example, the influence of product sales in winter and summer due to differences in temperature. Using the Seasonal Trend decomposition procedure based on Loss (STL), which is statistically used to process time data series, each historical data D _k in the historical data series {D _k } is divided into The value of the historical sales entity M _k of , is split into three parts, which are the component corresponding to the influence of seasonal factors, the component associated with the sales trend, and the white noise component. Among them, the value of the component corresponding to the influence of seasonal factors is deleted from the value of the historical sales entity _Mk , and only the values corresponding to the other two components are retained. It can be seen that in step 122, the historical data element _Dk is not removed using the STL method, but a part of the value of its sub-variable historical sales entity _Mk is deleted. That is to say, the number of elements of the historical data sequence processed by the STL method is not reduced, which is different from the process of removing the influence of holiday factors in step 121 . In traditional economics price sensitivity models, at least one of holiday factors and seasonal factors is generally not considered.

According to an embodiment of the present application, at least one of the part affected by holiday factors and the part affected by seasonal factors may be performed to generate processed historical data before inputting the historical data series {D _k } into the regression model sequence { _Dk ^* }.

In optional step 140, a logarithmic operation may be applied to the sales volume entity _Mk ^* and the price entity _Pk in the historical data _Dk ^* in the created processed historical data sequence { _Dk ^* }, respectively, eg Use the natural logarithm operation. The benefit of using logarithmic operations is to simplify price sensitivity calculations later on, which are described below.

The historical data sequence processed above is input into the regression model in step 150 to perform linear regression to complete the process of fitting the historical trend.

The input to the regression model, in addition to the processed historical data sequence { _Dk ^* }, also includes the set of trend polynomials provided in step 151 for fitting a predetermined sales trend, the set including at least one trend polynomial, each Trend polynomials all correspond to predetermined sales trends. A trend polynomial that fits a predetermined sales trend is a polynomial based on a time variable that provides a trend alternative to the regression model. The sales trends of stores are generally divided into three categories: rising, falling and fluctuating trends, and the change parameters in each type of sales trend are different. The sales trend polynomial is a linear relationship with the sales volume entity (sales volume data) of the product as the dependent variable and the price entity (price data) as the independent variable, so the regression model is used to determine the optimal trend that can accurately fit the sales trend of the store A polynomial to determine the price sensitivity defined by the ratio of the magnitude of changes in the sales volume and price entities. The order of the terms of the price entity in the sales trend polynomial can be between -0.25 and +4. The method used to determine price sensitivity models each store, so to accommodate the unique sales trends of each store, a collection of trend polynomials (i.e. pools of candidates) is provided for the method to automatically try to determine the appropriateness for the current store. The optimal trend polynomial of , and the coefficients of the correlation terms.

According to the embodiment of the present application, the regression model may be a linear regression model, and a step-wise regression linear model is used below to describe the solution of the present application. The stepwise regression linear model is a linear regression independent variable selection model. Each time a new independent variable is introduced, an F test is performed, and a t test is performed for the independent variables that have been selected one by one. When a previously introduced independent variable becomes no longer significant (or causes multicollinearity) due to the introduction of a later independent variable, the previously introduced independent variable is deleted to ensure that the regression is performed before each new independent variable is introduced. Only significant variables are included in the equation or polynomial. Stepwise regression of variable introduction and validation is performed repeatedly until neither significant independent variables are introduced into the regression equation or polynomial, and no insignificant independent variables are removed from the regression equation or polynomial. This can ensure that the finally obtained set of independent variables is optimal for the fitting of the regression equation or polynomial, that is, each independent variable in the obtained set of independent variables is significant relative to the regression equation or polynomial. In the stepwise regression model, an expected fitting target (such as a fitting error threshold) can also be set, so that when the expected fitting target cannot be met, it can be determined that the regression equation or polynomial cannot meet the requirements based on the current input set of independent variables. The best fit, or that the regression equation or polynomial is not the best fit equation or polynomial corresponding to the input set of independent variables.

In the regression model, the price entity of the historical data in the processed historical data sequence is used as a fixed independent variable, the trend polynomial corresponding to the predetermined sales trend is used as a variable independent variable, and the sales volume entity in the historical data is used as the dependent variable, Common input stepwise regression linear model. Each time a trend polynomial is selected from the trend polynomial set for step-by-step fitting, and all the historical data D _k ^* in the historical data sequence {D _k ^* } are sequentially introduced into the trend polynomial to test and judge that the historical data D _k ^* is relative to the trend polynomial. The significance of this trend polynomial. For regression operations, historical data can be traversed from front to back in chronological order, or historical data can be traversed from back to front. If all the historical data introduced into the stepwise regression linear model are removed from the significant data, the fitting of the retained significant historical data for the trend polynomial meets the expected fitting target, and it can be determined that the trend polynomial can be accurate to characterize the relationship between the sales volume entity and the price entity in the historical data sequence {D _k ^* }. The desired fit target may be, for example, that the error of fitting the significant historical data set to the trend polynomial is within the desired error threshold. If the retained historical data set cannot meet the expected fitting target for the trend polynomial, it means that the trend polynomial cannot accurately represent the relationship between the sales volume entity and the price entity in the historical data series {D _k ^* }. Select other trend polynomials in and repeat the above regression calculation process.

The price sensitivity determination method of the present application can also adopt other regression models, as long as the model can take the historical data in the processed historical data series as the fixed independent variable of the input regression model, and convert multiple trends corresponding to the predetermined sales trends The polynomial is used as the variable independent variable of the input regression model, and the sales entity in the historical data in the processed historical data series is used as the dependent variable of the input regression model, and the significant historical data is retained through the regression operation and determined to meet the expectations. The optimal trend polynomial for the target.

As shown in step 152, a temporal weight-based attention mechanism may also be added before the processed historical data sequence {D _k ^* } is input into the regression model. The attention mechanism is used to adjust the influence of the historical data D _k ^* corresponding to different historical time T _k ^* on the fitting trend polynomial in the regression operation. For example, historical sales volume data generated recently should be more representative of store sales trends than historical sales volume data generated relatively far back in time. The attention mechanism uses a polynomial based on the time variable to calculate the weights, adjusting the effect of historical time on the trend regression operation. For example, it can be ensured that the historical sales volume entity in the more recent historical data _Dk ^* has a higher weight. According to the embodiments of the present application, the weekly ordinal k of the historical time T _k in the historical time series {T _k ^* } is used as the time variable, and the polynomial is selected as the power of k, such as k 0.5 to the power k ^0.5 , then each historical The data Dk ^* is weighted with a power of _k (k ^0.5 ). The closer to the historical data of the current time, the larger its k value, and thus the larger its weight k ^0.5 .

The output of the regression model is the determined optimal trend polynomial corresponding to the sales trend of the product, and the price sensitivity can be determined according to the optimal trend polynomial. According to the definition of price sensitivity, the price sensitivity is related to the coefficient of the term in the optimal trend polynomial including the price entity of the historical data _Dk ^* . Since the logarithmic operation has been performed on the historical sales volume entity and the price entity in the historical data _Dk ^* in step 140, although the regression model uses the historical data _Dk ^* in the historical data series { _Dk ^* } and the trend polynomial Predicts store sales trends, but the coefficients of terms with price entities can be used directly to calculate price sensitivity. When the natural logarithm operation is used in step 140, the coefficient of the term including the price entity is effectively equivalent to the price sensitivity.

In the process of determining the optimal trend polynomial by the regression model, the historical data D _k ^* (1≤k≤m, m is retained after removing the insignificant historical data, which is included as a fixed independent variable in the optimal trend polynomial is also generated. The significance value p (0<p<1) of the coefficient corresponding to the item of the historical price entity in the number of historical data). The significance value p characterizes the reliability of using the coefficient corresponding to the term in the optimal trend polynomial including the historical price entity to accurately characterize the sales trend. For the coefficient of the item including the historical price entity as a fixed independent variable in the optimal trend polynomial, the weighted price sensitivity is obtained by weighting according to the significance value p corresponding to each historical data D _k ^* , as shown in Figure 1 Step 160 is described. The weight corresponding to the historical data with the significance value p may be a value calculated based on a polynomial of p, such as (1-p) ² . The price sensitivity weighting operation based on the significance coefficient p is used to penalize the insignificant coefficients, so that the calculated price sensitivity has higher significance and higher credibility.

For different stores, the above steps can be used to obtain the price sensitivity of the corresponding store from the store's historical sales data. The price sensitivity data of different stores can be encapsulated into a flexible sorting data table and provided to the company decision makers or related financial personnel who manage multiple stores. The data table may include, for example, store numbers and significance-weighted price sensitivity. For example, in the above exemplary embodiment, the price sensitivity is the coefficient of the weighted price entity term.

The calculated price sensitivity can be notified or distributed to the user, eg, by mail or message.

After obtaining the price sensitivity data of the store, users can implement corresponding price strategies in the store according to different price sensitivity situations. For example, in step 171, the price sensitivity of each store is sorted and a price strategy is applied. According to the embodiment, the larger the absolute value of the price sensitivity (weighted coefficient of the price entity item), the less sensitive the customers of the store are to the price, and the more suitable for implementing the price increase strategy. On the contrary, if the absolute value of price sensitivity is smaller, it indicates that customers are more sensitive to price and need to be carefully considered when implementing the price increase strategy.

Another way of applying price strategies based on price sensitivity data is presented in step 172 . This method extracts appropriate thresholds according to the price sensitivity distribution of all stores, and sets a hierarchical strategy according to the threshold interval where the price sensitivity of the stores is located to divide the stores into multiple groups. For example, a group with a higher threshold corresponds to the least price-sensitive stores, and a higher price increase strategy can be applied. Correspondingly, a group with a relatively low threshold corresponds to a store that is less sensitive to price, and this group of stores can apply a small price increase strategy. For a group with a central threshold, it means that stores that are more sensitive to price can consider maintaining the original price. Groups with lower thresholds indicate that customers in their stores are very price-sensitive and can suggest promotions.

FIG. 2 illustrates a method of determining price sensitivity according to an embodiment of the present application.

In step 210, a historical data sequence is generated based on the historical sales data of the product, and the historical data sequence composed of the historical data corresponds to the historical time sequence composed of the associated historical time. The determination methods for the two sub-variables of historical data, historical sales volume data (ie, historical sales volume entity) and historical price data (ie, historical price entity), have been described above.

In step 220, a processed historical data series is generated by removing the portion of the historical data series that is affected by periodic factors. Step 220 is used to preprocess the historical data sequence. The preprocessing includes removing historical data associated with a historical time corresponding to at least one of a lunar holiday and a solar holiday in the historical data sequence for holiday factors. The preprocessing includes removing the components affected by seasonal factors in the value of historical sales data in the historical data series based on the STL method for seasonal factors. Logarithmic operations can also be applied to the historical sales volume data and historical price data of the historical data before inputting the processed historical data series into the regression model to facilitate subsequent price sensitivity calculations. The method of the present application may also introduce an attention mechanism by weighting the historical sales volume data of the historical data based on the historical time associated with the historical data in the processed historical data sequence.

Next, an optimal trend polynomial corresponding to the sales trend of the product is determined using a regression model in step 230 based on the plurality of trend polynomials corresponding to the predetermined sales trends and the two inputs of the processed historical data series, and further based on The optimal trend polynomial determines price sensitivity. Specifically, the price entity in the historical data in the processed historical data sequence is used as a fixed independent variable, a plurality of trend polynomials corresponding to predetermined sales trends are used as variable independent variables, and the The sales volume entity in the historical data is used as the dependent variable, and the significant historical data is retained through the regression operation and the optimal trend polynomial that satisfies the desired fitting target is determined. After obtaining the optimal trend polynomial, the coefficients of terms including historical price data can be used to determine price sensitivity. When calculating price sensitivity, a significance coefficient can also be introduced to weight the coefficients of items including price data to improve the credibility of price sensitivity.

In optional step 240, for the obtained price sensitivities of a plurality of sales units, a price strategy may be applied according to the sorted results, or grouped based on the thresholds, and corresponding price strategies may be applied to different groups respectively.

FIG. 3 shows an apparatus for determining price sensitivity according to an embodiment of the present application. The device 300 includes at least a historical data sequence generating unit 310 , a historical data sequence processing unit 320 , a price sensitivity determining unit 330 and a price strategy generating unit 340 . The historical data sequence generating unit 310 is configured to generate a historical data sequence based on the historical sales data of the product. The unit 310 may also be used to implement the specific functions described in step 210 above. The historical data sequence processing unit 320 is configured to generate a processed historical data sequence by removing the part affected by periodic factors in the historical data sequence. Further, this unit 320 also implements other functions completed by step 220 in FIG. 2 . The price sensitivity determination unit 330 is configured to realize the two inputs completed in step 230 based on a plurality of trend polynomials corresponding to the predetermined sales trend and the processed historical data series, and use a regression model to determine the most suitable product corresponding to the sales trend of the product. The optimal trend polynomial, and further determine the price sensitivity based on the optimal trend polynomial. The price strategy generation unit 340 is configured to apply the price strategy according to the sorted results according to the obtained price sensitivities of the multiple sales units, and/or group them based on their thresholds and apply corresponding price strategies for different groups respectively.

According to the method and apparatus for determining price sensitivity of the present application, based on the regression algorithm model in machine learning instead of the traditional economic model considering price and sales, data preprocessing methods such as STL are used to convert historical data before inputting the regression model. Parts of the data series affected by periodic factors are removed, and a pool of trend polynomial candidates associated with historical time is added to the regression operation, using search-based automatic convergence methods such as stepwise regression linear models. In order to further consider the impact of historical time on price sensitivity, an attention mechanism is introduced to strengthen the fitting weight of recent historical data to avoid the dilution of information value caused by a long time line. The solution of the present application successfully fits external factors such as holiday and seasonal factors and store personalization trend factors to the maximum extent. Since the prices in the historical sales data used are not official pricing but a weighting of actual transaction prices, factors such as discounting activity are taken into account. The trend polynomial candidate pool can cover the personalized sales trends of rising, falling and fluctuating stores, including more influences of external factors on store sales. The price sensitivity between the price entity and the sales volume entity determined by the method and device is as close as possible to pure price sensitivity, controlling the impact of external factors on the historical sales volume entity, and the price sensitivity uses traditional economic models almost Unable to calculate. For multiple stores, the store price sensitivity ranking and grouping are ultimately used to guide the store's price strategy at the company level.

It should be noted that although several modules or units of the apparatus for determining price sensitivity are mentioned in the above detailed description, this division is not mandatory. Indeed, according to embodiments of the present application, the features and functions of two or more modules or units described above may be embodied in one module or unit. Conversely, the features and functions of one module or unit described above may be further divided into multiple modules or units to be embodied. Components shown as modules or units may or may not be physical units, ie may be located in one place, or may be distributed over multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of the present application. Those of ordinary skill in the art can understand and implement it without creative effort.

In an exemplary embodiment of the present application, there is also provided a computer-readable storage medium on which a computer program is stored, the program including executable instructions, which, when executed by, for example, a processor, can implement any one of the above The steps of the method for determining price sensitivity described in the Examples. In some possible implementations, various aspects of the present application can also be implemented in the form of a program product, which includes program code, which is used to cause the program product to run on a terminal device when the program product is executed. The terminal device performs the steps according to various exemplary embodiments of the present application described in the method for determining price sensitivity in this specification.

The program product for implementing the above method according to the embodiments of the present application may adopt a portable compact disc read only memory (CD-ROM) and include program codes, and may be executed on a terminal device such as a personal computer. However, the program product of the present application is not limited thereto, and in this document, a readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

The program product may employ any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus or device, or a combination of any of the above. More specific examples (non-exhaustive list) of readable storage media include: electrical connections with one or more wires, portable disks, hard disks, random access memory (RAM), read only memory (ROM), erasable programmable read only memory (EPROM or flash memory), optical fiber, portable compact disk read only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

The computer-readable storage medium may include a data signal propagated in baseband or as part of a carrier wave, carrying readable program code therein. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the foregoing. A readable storage medium can also be any readable medium other than a readable storage medium that can transmit, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device. Program code embodied on a readable storage medium may be transmitted using any suitable medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

Program code for carrying out the operations of the present application may be written in any combination of one or more programming languages, including object-oriented programming languages—such as Java, C++, etc., as well as conventional procedural Programming Language - such as the "C" language or similar programming language. The program code may execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device and partly on a remote computing device, or entirely on the remote computing device or server execute on. In the case of a remote computing device, the remote computing device may be connected to the user computing device through any kind of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (eg, using an Internet service provider business via an Internet connection).

In an exemplary embodiment of the present application, there is also provided an electronic device, which may include a processor, and a memory for storing executable instructions of the processor. Wherein, the processor is configured to perform the steps of the method for determining price sensitivity in any one of the above embodiments by executing the executable instructions.

As will be appreciated by one skilled in the art, various aspects of the present application may be implemented as a system, method or program product. Therefore, various aspects of the present application can be embodied in the following forms, namely: a complete hardware implementation, a complete software implementation (including firmware, microcode, etc.), or a combination of hardware and software aspects, which may be collectively referred to herein as implementations "circuit", "module" or "system".

The electronic device 400 according to this embodiment of the present application is described below with reference to FIG. 4 . The electronic device 400 shown in FIG. 4 is only an example, and should not impose any limitations on the functions and scope of use of the embodiments of the present application.

As shown in FIG. 4, electronic device 400 takes the form of a general-purpose computing device. Components of the electronic device 400 may include, but are not limited to, at least one processing unit 410, at least one storage unit 420, a bus 430 connecting different system components (including the storage unit 420 and the processing unit 410), a display unit 440, and the like.

Wherein, the storage unit stores program codes, and the program codes can be executed by the processing unit 410, so that the processing unit 410 executes various examples according to the present application described in the method for determining price sensitivity in this specification steps of sexual implementation. For example, the processing unit 410 may perform the steps shown in FIG. 1 and FIG. 2 .

The storage unit 420 may include a readable medium in the form of a volatile storage unit, such as a random access storage unit (RAM) 4201 and/or a cache storage unit 4202 , and may further include a read only storage unit (ROM) 4203 .

The storage unit 420 may also include a program/utility 4204 having a set (at least one) of program modules 4205 including, but not limited to, an operating system, one or more application programs, other program modules, and programs Data, each or some combination of these examples may include an implementation of a network environment.

The bus 430 may be representative of one or more of several types of bus structures, including a memory cell bus or memory cell controller, a peripheral bus, a graphics acceleration port, a processing unit, or a local area using any of a variety of bus structures. bus.

The electronic device 400 may also communicate with one or more external devices 500 (eg, keyboards, pointing devices, Bluetooth devices, etc.), with one or more devices that enable a user to interact with the electronic device 400, and/or with Any device (eg, router, modem, etc.) that enables the electronic device 400 to communicate with one or more other computing devices. Such communication may occur through input/output (I/O) interface 450 . Also, the electronic device 400 may communicate with one or more networks (eg, a local area network (LAN), a wide area network (WAN), and/or a public network such as the Internet) through a network adapter 460 . Network adapter 460 may communicate with other modules of electronic device 400 through bus 430 . It should be understood that, although not shown, other hardware and/or software modules may be used in conjunction with electronic device 400, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives and data backup storage systems.

From the description of the above embodiments, those skilled in the art can easily understand that the exemplary embodiments described herein may be implemented by software, or may be implemented by software combined with necessary hardware. Therefore, the technical solutions according to the embodiments of the present application may be embodied in the form of software products, and the software products may be stored in a non-volatile storage medium (which may be CD-ROM, U disk, mobile hard disk, etc.) or on the network , including several instructions to cause a computing device (which may be a personal computer, a server, or a network device, etc.) to execute the method for determining price sensitivity according to an embodiment of the present application.

Other embodiments of the present application will readily occur to those skilled in the art upon consideration of the specification and practice of what is disclosed herein. This application is intended to cover any variations, uses or adaptations of this application that follow the general principles of this application and include common knowledge or conventional techniques in the technical field not disclosed in this application . The specification and examples are to be regarded as exemplary only, with the true scope and spirit of the application being indicated by the appended claims.

Claims

A method for determining price sensitivity, comprising:

A historical data series is generated based on historical sales data of the product, the historical data series is composed of historical data associated with historical times in the historical time series, wherein the historical data includes historical sales volume data and historical price of the product data;

generating a processed historical data series by removing the portion of the historical data series that is affected by periodic factors;

Based on a plurality of trend polynomials corresponding to predetermined sales trends and the processed historical data series, a regression model is used to determine an optimal trend polynomial corresponding to the sales trend of the product, and the optimal trend polynomial is determined based on the optimal trend polynomial. and the price sensitivity, wherein the sales volume data of the product is represented as the trend polynomial with the price data as an independent variable.
The method according to claim 1, wherein the historical sales volume data is determined based on the total sales volume of the product, and the historical price data is determined based on a weighted determination of the sales volume and the sales price of the product.
The method according to claim 1, wherein the periodic factors include holiday factors and seasonal factors.
The method according to claim 3, wherein the holiday factor includes a lunar calendar holiday and a solar calendar holiday, and by removing the historical data sequence corresponding to at least one of the lunar calendar holiday and the solar calendar holiday the historical data associated with historical time to generate the processed series of historical data.
The method according to claim 3 or 4, characterized in that, components affected by seasonal factors in the value of the historical sales volume data in the historical data are separated by a loss-based seasonal trend decomposition process (STL). removed to generate the processed series of historical data.
3. The method of claim 2, wherein applying to the historical sales volume data and the historical price data of the historical data prior to inputting the processed series of historical data into the regression model Logarithmic operations.
The method of claim 1, wherein prior to inputting the processed series of historical data into the regression model, based on a The historical time weights the historical sales volume data of the historical data.
The method of claim 1, wherein using a regression model to determine the optimal trend polynomial corresponding to the sales trend of the product further comprises:

The price entity in the historical data in the processed historical data sequence is used as a fixed independent variable input to the regression model, and a plurality of trend polynomials corresponding to predetermined sales trends are used as the variable input to the regression model. Change the independent variable, take the sales volume entity in the historical data in the processed historical data sequence as the dependent variable of the input to the regression model, retain the significant historical data through regression operation and determine to meet expectations Fits the optimal trend polynomial of the target.
The method according to claim 8, wherein the regression model is a stepwise regression linear model.
2. The method of claim 1, wherein determining the price sensitivity based on the optimal trend polynomial further comprises determining the price sensitivity based on a coefficient through a term in the optimal trend polynomial that includes the price data price sensitivity.
11. The method of claim 10, wherein determining the price sensitivity based on the optimal trend polynomial further comprises including the price based on a pair of significance coefficients associated with coefficients of items including the price data The coefficients of the terms of the data are weighted.
The method of claim 1, comprising determining corresponding price sensitivities for historical sales data for a plurality of sales units.
13. The method of claim 12, further comprising ranking the plurality of sales units based on respective price sensitivities of the plurality of sales units, and applying a price strategy based on the ranking results.
13. The method of claim 12, further comprising grouping the plurality of sales units based on corresponding price sensitivities of the plurality of sales units, and applying corresponding price policies for different groupings.
An apparatus for determining price sensitivity, comprising:

A historical data sequence generating unit configured to generate a historical data sequence based on historical sales data of the product, the historical data sequence being composed of historical data associated with historical times in the historical time series, wherein the historical data includes the Historical sales volume data and historical price data of the product;

a historical data sequence processing unit configured to generate a processed historical data sequence by removing a portion of the historical data sequence affected by periodic factors;

a price sensitivity determination unit configured to use a regression model to determine an optimal trend polynomial corresponding to a sales trend of the product based on a plurality of trend polynomials corresponding to predetermined sales trends and the processed historical data sequence, and The price sensitivity is determined based on the optimal trend polynomial, wherein the sales volume data for the product is represented as the trend polynomial with the price data as an independent variable.
A computer-readable storage medium having stored thereon a computer program comprising executable instructions which, when executed by a processor, implement the method of any one of claims 1 to 14.
An electronic device, comprising:

processor; and

a memory for storing executable instructions for the processor;

wherein the processor is arranged to execute the executable instructions to implement the method of any one of claims 1 to 14.