Bulk commodity price index prediction system based on multiple factors
A commodity price index prediction system predicts prices by integrating multiple factors like commodity variety, production areas, and weather, improving forecasting accuracy and adaptability.
Patent Information
- Application Number
- CN202411250087.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-06
- Publication Date
- 2024-12-06
AI Technical Summary
Existing commodity price forecasting systems fail to accurately predict commodity prices due to the lack of flexibility and adaptability, and the lack of flexibility and accuracy when processing and analyzing large-scale data, and the lack of flexibility and accuracy when processing and analyzing large-scale data, and the lack of flexibility and flexibility when processing and analyzing large-scale data.
A commodity price index prediction system utilizing a data processing module to predict commodity prices based on multiple factors.
The system accurately predicts commodity prices by considering various factors such as commodity variety, production areas, sales areas, and weather, enhancing flexibility and accuracy in forecasting.
Smart Images

Figure CN119090548A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of commodity price forecasting, and in particular to a commodity price index forecasting system based on multiple factors. Background Art
[0002] Commodities play a vital role in the global economic system. These commodities are usually traded globally, and their price fluctuations have a profound impact on the world economy. With the development of globalization, the prices of commodities are affected by more and more factors, so models that can accurately predict commodity prices are extremely valuable to market participants.
[0003] Traditional commodity price indices usually rely on a single supply and demand relationship or a simple historical price trend analysis, which is incapable of dealing with complex and changing market realities. In addition, most existing models fail to fully consider regional differences. For example, logistics costs and market demand in different regions may have significantly different effects on the price of the same commodity. Similarly, variety characteristics are often ignored, and the prices of different varieties of commodities in the market may vary greatly. Existing models have problems with efficiency and accuracy when processing and analyzing large-scale complex data, and these models often lack the necessary flexibility and adaptability. Therefore, there is an urgent need for a commodity price index prediction system based on multiple factors to achieve automated prediction of commodity price indices through intelligent data processing. Summary of the invention
[0004] The purpose of the present invention is to provide a commodity price index prediction system based on multiple factors to solve the problems that traditional methods usually rely on a single supply and demand relationship, cannot adapt to complex and changing markets, and have efficiency and accuracy problems when processing and analyzing large-scale complex data.
[0005] The present invention provides a commodity price index prediction system based on multiple factors, comprising:
[0006] A data acquisition module, used to acquire commodity price data and commodity forecast influencing factor data, wherein the forecast influencing factor data includes commodity variety data, commodity production area data, commodity sales area data and commodity sales area weather data;
[0007] A data processing module is used to pre-process the forecast influencing factor data, convert non-numeric data in the forecast influencing factor data into numeric data; and perform correlation analysis on the pre-processed forecast influencing factor data based on the price data to determine the key influencing factors of the commodity;
[0008] A model building module, used to set the weights of the key influencing factors and build a commodity price index model;
[0009] A price prediction module is used to predict the commodity price index according to the commodity price index model.
[0010] In some embodiments of the present application, the data processing module, when used to preprocess the prediction influencing factor data and convert non-numeric data in the prediction influencing factor data into numeric data, includes:
[0011] Determining the supply and demand data of the bulk commodity according to the bulk commodity variety data and the bulk commodity sales location data;
[0012] Determine the origin-sales location nonlinear coefficient data of the bulk commodity according to the origin data of the bulk commodity and the sales location data of the bulk commodity;
[0013] Determine weather temperature data when the bulk commodity is sold based on weather data of the bulk commodity sales location;
[0014] Obtain influencing factor data; wherein, the influencing factors include the supply and demand of bulk commodities, the non-linear coefficient of the production place and the sales place of bulk commodities, and the weather temperature when the bulk commodities are sold.
[0015] In some embodiments of the present application, when the data processing module is used to determine the supply and demand data of the bulk commodity according to the bulk commodity variety data and the bulk commodity sales location data, it includes:
[0016] Obtain the commodity demand types and commodity demand quantities in commodity sales areas;
[0017] Sort the commodity demand categories from large to small according to the commodity demand quantity;
[0018] Determine the ranking of the commodities according to the commodities;
[0019] Determining the supply and demand of the bulk commodity according to the ranking of the bulk commodity;
[0020] The supply and demand of the commodities are calculated according to the following formula:
[0021]
[0022] Among them, N D Expressed as the supply and demand of commodities, P D Expressed as the ranking of commodities, P Z The total number of commodity demand categories ranked by commodity sales location.
[0023] In some embodiments of the present application, when the data processing module is used to determine the origin-sales location nonlinear coefficient data of the bulk commodity according to the origin data of the bulk commodity and the sales location data of the bulk commodity, it includes:
[0024] Determine the upward and downward distances between commodity production areas and commodity sales areas, as well as the shortest road distance between commodity production areas and commodity sales areas;
[0025] Determine the production-sales location nonlinear coefficient of the bulk commodity according to the upward distance, the downward distance and the shortest road distance;
[0026] The nonlinear coefficient between production location and sales location is calculated according to the following formula:
[0027]
[0028] Among them, R D It is expressed as the non-linear coefficient of production place-sales place, L1 is expressed as the upward distance, L2 is expressed as the downward distance, and D is expressed as the shortest road distance.
[0029] In some embodiments of the present application, when the data processing module is used to determine the weather temperature data when the bulk commodity is sold according to the weather data of the bulk commodity sales area, it includes:
[0030] Dividing the weather data of the bulk commodity sales areas according to seasons, and classifying the weather data of the bulk commodity sales areas in each season to obtain weather categories;
[0031] Determine a weather cycle according to the weather category, wherein the weather cycle includes a cold cycle and a warm cycle;
[0032] The weather temperature at the time of commodity sales is determined based on the weather cycle and temperature values.
[0033] In some embodiments of the present application, when the data processing module is used to determine the weather temperature when the bulk commodity is sold according to the weather cycle and the temperature value, it includes:
[0034] Determine a corresponding weather coefficient according to the weather cycle;
[0035] When the weather cycle is a cold cycle, the weather coefficient is selected as the first weather coefficient, and when the weather cycle is a warm cycle, the weather coefficient is selected as the second weather coefficient;
[0036] Determine the weather temperature when the commodity is sold based on the weather coefficient and temperature value;
[0037] The weather temperature at the time of commodity sales is calculated according to the following formula:
[0038] TD =A+BT W ;
[0039] Among them, T D It is expressed as the weather temperature when the commodity is sold, A is the seasonal coefficient, B is the weather coefficient, T W Expressed as a temperature value.
[0040] In some embodiments of the present application, when the data processing module is used to perform a correlation analysis on the pre-processed forecast influencing factor data based on the price data to determine the key influencing factors of the commodity, it includes:
[0041] SPSS software was used to conduct significance analysis and correlation analysis on the preprocessed prediction influencing factor data;
[0042] Input the commodity price data and influencing factor data into SPSS software, select bivariate analysis, Pearson correlation coefficient and two-tailed significance test, and obtain the significance level and Pearson correlation coefficient of the influencing factors;
[0043] If the significance level is less than or equal to 5%, it is determined that there is a significant linear correlation between the influencing factor corresponding to the significance level and the commodity price;
[0044] If the significance level is greater than 5%, it is determined that there is no significant linear correlation between the influencing factor corresponding to the significance level and the commodity price;
[0045] If there is a significant linear correlation between the influencing factor corresponding to the significance level and the commodity price, whether the influencing factor is a key influencing factor is determined based on the Pearson correlation coefficient.
[0046] In some embodiments of the present application, when the data processing module is used to determine whether the influencing factor is a key influencing factor according to the Pearson correlation coefficient, it includes:
[0047] There is a preset Pearson correlation coefficient;
[0048] Comparing the Pearson correlation coefficient with the preset Pearson correlation coefficient;
[0049] If the Pearson correlation coefficient is greater than or equal to the preset Pearson correlation coefficient, determining that the influencing factor corresponding to the Pearson correlation coefficient is a key influencing factor;
[0050] If the Pearson correlation coefficient is less than the preset Pearson correlation coefficient, it is determined that the influencing factor corresponding to the Pearson correlation coefficient is not a key influencing factor.
[0051] In some embodiments of the present application, the model building module, when used to set the weights of the key influencing factors and generate a commodity price index model, includes:
[0052] Establish a commodity price index forecasting model based on the key influencing factors;
[0053] Training the commodity price index prediction model according to the influencing factor data to obtain multiple sets of coefficient values;
[0054] Determining weights of the key influencing factors according to the multiple groups of coefficient values, fitting the weights with the commodity price index prediction model, and generating the commodity price index model;
[0055] The expression of the commodity price index model is:
[0056] K=αN D +βR D +γT D ;
[0057] Among them, α represents the weight of the supply and demand of commodities, N D It is expressed as the supply and demand of commodities, β is expressed as the weight of the nonlinear coefficient of the production and sales location of commodities, R D It is expressed as the nonlinear coefficient of production place-sales place, γ is the weight of weather temperature when commodities are sold, T D Indicates how cold or warm the weather is when the commodity is sold.
[0058] In some embodiments of the present application, when the price prediction module is used to predict the commodity price index according to the commodity price index model, it includes:
[0059] Obtain data on key influencing factors;
[0060] Based on the commodity price index model, the commodity price index is predicted according to the key influencing factor data to obtain a predicted value of the commodity price index.
[0061] Compared with the prior art, the beneficial effect of the present invention is that, through the data acquisition module, a wealth of commodity price data and a variety of forecast influencing factor data can be collected, ensuring the comprehensiveness and accuracy of the forecast. The data processing module converts non-numerical data, making the data easier to analyze and process, laying the foundation for subsequent correlation analysis and model building. Correlation analysis based on price data can accurately find out the factors that have an important impact on commodity prices and improve the pertinence of the forecast. The model building module sets the weights of key influencing factors to establish a reasonable price index model, making the forecast more scientific and reliable. The price forecast module uses the established model for forecasting, which can provide a relatively accurate forecast result of the commodity price index and help relevant decision-making. The system can adapt to changes in factors such as different commodity varieties, production areas, sales areas, and weather, and has good adaptability and flexibility; it provides valuable reference information for market participants, helping them make more informed decisions. BRIEF DESCRIPTION OF THE DRAWINGS
[0062] 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 in the following description are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.
[0063] Figure 1 It is a functional block diagram of a commodity price index prediction system based on multiple factors of the present invention. DETAILED DESCRIPTION
[0064] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.
[0065] like Figure 1 As shown, the present invention provides a commodity price index prediction system based on multiple factors, including:
[0066] The data acquisition module is used to acquire commodity price data and forecast influencing factor data of commodities, wherein the forecast influencing factor data includes commodity variety data, commodity origin data, commodity sales location data and commodity sales location weather data.
[0067] The data processing module is used to pre-process the forecast influencing factor data, convert the non-numerical data in the forecast influencing factor data into numerical data; and perform correlation analysis on the pre-processed forecast influencing factor data based on the price data to determine the key influencing factors of commodities.
[0068] The model building module is used to set the weights of the key influencing factors and build a commodity price index model.
[0069] A price prediction module is used to predict the commodity price index according to the commodity price index model.
[0070] In some embodiments of the present application, the data processing module, when used to preprocess the prediction influencing factor data and convert the non-numeric data in the prediction influencing factor data into numerical data, includes: determining the supply and demand data of the commodities based on the commodity variety data and the commodity sales location data; determining the production location-sales location non-linear coefficient data of the commodities based on the commodity production location data and the commodity sales location data; determining the weather warmth data when the commodities are sold based on the weather data of the commodity sales location; and obtaining the influencing factor data; wherein the influencing factors include the supply and demand of the commodities, the production location-sales location non-linear coefficient of the commodities and the weather warmth when the commodities are sold.
[0071] In this embodiment, most of the forecast influencing factor data obtained are non-numerical data, such as bulk commodity variety data, bulk commodity origin data, and bulk commodity sales location data, where variety indicates the type, origin or sales location indicates the geographical location, and bulk commodity sales location weather data also contains some non-numerical data, such as weather categories such as sunny, rainy, cloudy, etc. Therefore, it is necessary to pre-process the forecast influencing factor data and convert the non-numerical data therein into numerical data, such as bulk commodity origin data and bulk commodity sales location data, which can reflect the transportation convenience and transportation cost between two places; bulk commodity variety data and bulk commodity sales location data can reflect the demand for bulk commodities; and weather data can reflect the impact of cold and warm weather on bulk commodities.
[0072] In some embodiments of the present application, when the data processing module is used to determine the supply and demand data of commodities based on the commodity variety data and the commodity sales location data, it includes: obtaining the commodity demand types and commodity demand quantities at the commodity sales locations; sorting the commodity demand types from large to small according to the commodity demand quantities; determining the corresponding rankings of the commodities according to the commodity varieties; and determining the supply and demand of the commodities according to the rankings of the commodities.
[0073] Among them, the supply and demand of commodities are calculated according to the following formula:
[0074]
[0075] Among them, N D Expressed as the supply and demand of commodities, P D Expressed as the ranking of commodities, P Z The total number of commodity demand categories ranked by commodity sales location.
[0076] In this embodiment, bulk commodities usually refer to raw materials traded globally, such as energy, metals, and agricultural products. These commodities are not only the basis of industrial and agricultural production, but also an important part of the financial investment field. There are differences in the demand for bulk commodities in different regions, and the differences in the demand for bulk commodities in various regions reflect the comprehensive effects of various factors such as their economic development level, industrial structure, population size, and resource endowment. For example, the differences between developed and developing countries in terms of industrialization, infrastructure construction, and per capita income level lead to significant differences in their demand for bulk commodities and demand structure; the leading industries in different regions are different, such as some countries may be dominated by agriculture, while other countries are dominated by heavy industry or service industry, which directly affects the type and quantity of demand for bulk commodities; regions with large populations have a greater total consumption of food, housing, and other basic needs, and thus have a higher demand for related bulk commodities. Therefore, different regions have different demands for bulk commodities. The present invention determines the ranking of the bulk commodity in the sales area by sorting all the bulk commodity demands in the sales area, and determines the supply and demand of the sales area for the bulk commodity according to the proportion of the ranking in the total ranking.
[0077] In some embodiments of the present application, the data processing module, when used to determine the non-linear coefficient data of the place of origin-sales location of the bulk commodity based on the place of origin data of the bulk commodity and the place of sales location of the bulk commodity, includes: determining the upward distance and downward distance between the place of origin of the bulk commodity and the place of sales of the bulk commodity, and the shortest road distance between the place of origin of the bulk commodity and the place of sales of the bulk commodity; determining the non-linear coefficient of the place of origin-sales location of the bulk commodity based on the upward distance, the downward distance and the shortest road distance;
[0078] The nonlinear coefficient between production location and sales location is calculated according to the following formula:
[0079]
[0080] Among them, R D It is expressed as the non-linear coefficient of production place-sales place, L1 is expressed as the upward distance, L2 is expressed as the downward distance, and D is expressed as the shortest road distance.
[0081] In this embodiment, the non-linear coefficient refers to the ratio of the actual traffic distance between the starting and ending points of the road to the spatial straight-line distance between the two points. It can evaluate the convenience of the connection between different road network types and passenger and freight flow route distribution points. The present invention uses an improved non-linear coefficient to represent the non-linear coefficient of the place of production-sales location. The length in the upward and downward directions refers to the total distance traveled in two directions from the starting station to the terminal station and then from the terminal station back to the starting station in a route. That is to say, the upward distance L1 is the distance traveled by the vehicle from the place of production to the sales location, and the downward distance L2 is the distance traveled by the vehicle from the sales location to the place of production. The denominator in the improved calculation formula of the present invention is the shortest road distance between the place of production and the sales location, rather than the spatial straight-line distance. The non-linear coefficient of the place of production-sales location calculated in this way is more statistically significant.
[0082] In some embodiments of the present application, when the data processing module is used to determine the weather temperature data when the bulk commodities are sold based on the weather data of the bulk commodity sales areas, it includes: dividing the weather data of the bulk commodity sales areas according to seasons, and classifying the weather data of the bulk commodity sales areas in each season to obtain weather categories; determining the weather cycle based on the weather category, the weather cycle including cold cycles and warm cycles; determining the weather temperature when the bulk commodities are sold based on the weather cycle and the temperature value.
[0083] In some embodiments of the present application, when the data processing module is used to determine the weather warmth when bulk commodities are sold based on the weather cycle and the temperature value, it includes: determining the corresponding weather coefficient based on the weather cycle; when the weather cycle is a cold cycle, selecting the weather coefficient as the first weather coefficient, and when the weather cycle is a warm cycle, selecting the weather coefficient as the second weather coefficient; determining the weather warmth when bulk commodities are sold based on the weather coefficient and the temperature value.
[0084] The weather temperature at the time of commodity sales is calculated according to the following formula:
[0085] T D =A+BT W ;
[0086] Among them, T D It is expressed as the weather temperature when the commodity is sold, A is the seasonal coefficient, B is the weather coefficient, T W Expressed as a temperature value.
[0087] In this embodiment, the weather data of the commodity sales area is long-term data. By analyzing the weather data of the sales area, it can be determined whether the change in weather temperature has an impact on the commodity price. First, the weather data is divided according to the season, and the divided weather data is classified to determine the weather category, such as determining whether the statistical day is sunny, rainy, snowy or cloudy, and determining the weather cycle according to the weather category, such as setting sunny days as warm cycles and rainy days as cold cycles. When determining the weather temperature, first determine the seasonal coefficient and the weather coefficient, where the seasonal coefficient can be determined according to the longitude and latitude of the sales area. If the sales area is in the northern hemisphere, it can be determined that winter is from December to February. If the sales area is in the southern hemisphere, it can be determined that winter is from June to September. Therefore, the season is determined by the longitude and latitude of the sales area and the time, and different seasonal coefficients are determined according to the season, such as the seasonal coefficient of winter is 0 and the seasonal coefficient of summer is 50. Weather temperature T D The smaller it is, the colder it is. D The bigger it is, the warmer it is.
[0088] In some embodiments of the present application, when the data processing module is used to perform correlation analysis on the pre-processed predicted influencing factor data based on price data to determine the key influencing factors of commodities, it includes: using SPSS software to perform significance analysis and correlation analysis on the pre-processed predicted influencing factor data; inputting the commodity price data and the influencing factor data into the SPSS software, selecting bivariate analysis, Pearson correlation coefficient and two-tailed significance test, and obtaining the significance level and Pearson correlation coefficient of the influencing factor; if the significance level is less than or equal to 5%, it is determined that there is a significant linear correlation between the influencing factor corresponding to the significance level and the commodity price; if the significance level is greater than 5%, it is determined that there is no significant linear correlation between the influencing factor corresponding to the significance level and the commodity price; if there is a significant linear correlation between the influencing factor corresponding to the significance level and the commodity price, then determining whether the influencing factor is a key influencing factor based on the Pearson correlation coefficient.
[0089] In this embodiment, SPSS is a widely used statistical analysis software with user-friendly data management capabilities and powerful statistical analysis functions. The present invention can directly obtain the significance level and correlation coefficient between the commodity price and the influencing factors by using SPSS software, determine whether there is a significant linear relationship between the influencing factors and the commodity price according to the significance level, and determine the correlation between the influencing factors and the commodity price according to the correlation coefficient when there is a significant linear correlation, and then determine the key influencing factors.
[0090] In some embodiments of the present application, when the data processing module is used to determine whether an influencing factor is a key influencing factor based on the Pearson correlation coefficient, it includes: setting a preset Pearson correlation coefficient; comparing the Pearson correlation coefficient with the preset Pearson correlation coefficient; if the Pearson correlation coefficient is greater than or equal to the preset Pearson correlation coefficient, determining that the influencing factor corresponding to the Pearson correlation coefficient is a key influencing factor; if the Pearson correlation coefficient is less than the preset Pearson correlation coefficient, determining that the influencing factor corresponding to the Pearson correlation coefficient is not a key influencing factor.
[0091] In this embodiment, by comparing the Pearson correlation coefficient with the preset Pearson correlation coefficient, it is determined whether the corresponding influencing factor is a key influencing factor according to the comparison result.
[0092] In some embodiments of the present application, when the model building module is used to set the weights of key influencing factors and generate a commodity price index model, it includes: establishing a commodity price index prediction model based on the key influencing factors; training the commodity price index prediction model based on the influencing factor data to obtain multiple groups of coefficient values; determining the weights of the key influencing factors based on the multiple groups of coefficient values, fitting the weights with the commodity price index prediction model, and generating a commodity price index model.
[0093] The expression of the commodity price index model is:
[0094] K=αN D +βR D +γT D ;
[0095] Among them, α represents the weight of the supply and demand of commodities, N D It is expressed as the supply and demand of commodities, β is expressed as the weight of the nonlinear coefficient of the production and sales location of commodities, R D It is expressed as the nonlinear coefficient of production place-sales place, γ is the weight of weather temperature when commodities are sold, T D Indicates how cold or warm the weather is when the commodity is sold.
[0096] In this embodiment, when the key influencing factors have been determined, a commodity price index prediction model is established, which includes the coefficient values of each key influencing factor, that is, the weights of each key influencing factor. By solving the commodity price index prediction model, multiple sets of coefficient values can be obtained, and weights are determined according to these coefficient values, thereby generating a commodity price index model.
[0097] In some embodiments of the present application, when the price prediction module is used to predict the commodity price index according to the commodity price index model, it includes: obtaining key influencing factor data; based on the commodity price index model, predicting the commodity price index according to the key influencing factor data, to obtain a predicted value of the commodity price index.
[0098] In this embodiment, when predicting the commodity price index, the key influencing factor data is first obtained, and the key influencing factor data is input into the commodity price index model to obtain the corresponding predicted value of the commodity price index.
[0099] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solution of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solution to deviate from the spirit and scope of the technical solution of the present invention.
[0100] The system provided in the above embodiment is only illustrated by the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the modules or steps in the embodiments of the present invention can be decomposed or combined. For example, the modules in the above embodiment can be combined into one module, or further divided into multiple sub-modules to complete all or part of the functions described above. The names of the modules and steps involved in the embodiments of the present invention are only for distinguishing the modules or steps, and are not regarded as improper limitations of the present invention.
[0101] Those skilled in the art should be able to appreciate that the modules and method steps of each example described in conjunction with the embodiments disclosed herein can be implemented with electronic hardware, computer software or a combination of the two, and the programs corresponding to the software modules and method steps can be placed in random access memory (RAM), memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disks, removable disks, CD-ROMs or any other form of storage medium known in the technical field. In order to clearly illustrate the interchangeability of electronic hardware and software, the composition and steps of each example have been generally described in the above description according to the function. Whether these functions are performed in electronic hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.
Claims
1. A commodity price index prediction system based on multiple factors, characterized in that: include: A data acquisition module, used to acquire commodity price data and commodity forecast influencing factor data, wherein the forecast influencing factor data includes commodity variety data, commodity production area data, commodity sales area data and commodity sales area weather data; A data processing module is used to pre-process the forecast influencing factor data, convert non-numeric data in the forecast influencing factor data into numeric data; and perform correlation analysis on the pre-processed forecast influencing factor data based on the price data to determine the key influencing factors of the commodity; A model building module, used to set the weights of the key influencing factors and build a commodity price index model; A price prediction module is used to predict the commodity price index according to the commodity price index model.
2. The commodity price index prediction system based on multiple factors according to claim 1 is characterized in that: The data processing module, when used to pre-process the prediction influencing factor data and convert non-numeric data in the prediction influencing factor data into numeric data, includes: Determining the supply and demand data of the bulk commodity according to the bulk commodity variety data and the bulk commodity sales location data; Determine the origin-sales location nonlinear coefficient data of the bulk commodity according to the origin data of the bulk commodity and the sales location data of the bulk commodity; Determine weather temperature data when the bulk commodity is sold based on weather data of the bulk commodity sales location; Obtain influencing factor data; wherein, the influencing factors include the supply and demand of bulk commodities, the non-linear coefficient of the production place and the sales place of bulk commodities, and the weather temperature when the bulk commodities are sold.
3. The commodity price index prediction system based on multiple factors according to claim 2 is characterized in that: When the data processing module is used to determine the supply and demand data of the bulk commodity according to the bulk commodity variety data and the bulk commodity sales location data, it includes: Obtain the commodity demand types and commodity demand quantities in commodity sales areas; Sort the commodity demand categories from large to small according to the commodity demand quantity; Determine the ranking of the commodities according to the commodities; Determining the supply and demand of the bulk commodity according to the ranking of the bulk commodity; The supply and demand of the commodities are calculated according to the following formula: Among them, N D Expressed as the supply and demand of commodities, P D Expressed as the ranking of commodities, P Z The total number of commodity demand categories ranked by commodity sales location.
4. The commodity price index prediction system based on multiple factors according to claim 2 is characterized in that: When the data processing module is used to determine the origin-sales location nonlinear coefficient data of the bulk commodity according to the origin data of the bulk commodity and the sales location data of the bulk commodity, it includes: Determine the upward and downward distances between commodity production areas and commodity sales areas, as well as the shortest road distance between commodity production areas and commodity sales areas; Determine the production-sales location nonlinear coefficient of the bulk commodity according to the upward distance, the downward distance and the shortest road distance; The nonlinear coefficient between production location and sales location is calculated according to the following formula: Among them, R D It is expressed as the non-linear coefficient of production place-sales place, L1 is expressed as the upward distance, L2 is expressed as the downward distance, and D is expressed as the shortest road distance.
5. The commodity price index prediction system based on multiple factors according to claim 2 is characterized in that: The data processing module, when used to determine the weather temperature data when the bulk commodity is sold according to the weather data of the bulk commodity sales area, includes: Dividing the weather data of the bulk commodity sales areas according to seasons, and classifying the weather data of the bulk commodity sales areas in each season to obtain weather categories; Determine a weather cycle according to the weather category, wherein the weather cycle includes a cold cycle and a warm cycle; The weather temperature at the time of commodity sales is determined based on the weather cycle and temperature values.
6. The commodity price index prediction system based on multiple factors according to claim 5 is characterized in that: The data processing module, when used to determine the weather temperature when the bulk commodity is sold according to the weather cycle and the temperature value, includes: Determine a corresponding weather coefficient according to the weather cycle; When the weather cycle is a cold cycle, the weather coefficient is selected as the first weather coefficient, and when the weather cycle is a warm cycle, the weather coefficient is selected as the second weather coefficient; Determine the weather temperature when the commodity is sold based on the weather coefficient and temperature value; The weather temperature at the time of commodity sales is calculated according to the following formula: T D =A+BT W ; Among them, T D It is expressed as the weather temperature when the commodity is sold, A is the seasonal coefficient, B is the weather coefficient, T W Expressed as a temperature value.
7. The commodity price index prediction system based on multiple factors according to claim 2 is characterized in that: The data processing module, when used to perform correlation analysis on the pre-processed forecast influencing factor data based on the price data to determine the key influencing factors of the commodity, includes: SPSS software was used to conduct significance analysis and correlation analysis on the preprocessed prediction influencing factor data; Input the commodity price data and influencing factor data into SPSS software, select bivariate analysis, Pearson correlation coefficient and two-tailed significance test, and obtain the significance level and Pearson correlation coefficient of the influencing factors; If the significance level is less than or equal to 5%, it is determined that there is a significant linear correlation between the influencing factor corresponding to the significance level and the commodity price; If the significance level is greater than 5%, it is determined that there is no significant linear correlation between the influencing factor corresponding to the significance level and the commodity price; If there is a significant linear correlation between the influencing factor corresponding to the significance level and the commodity price, whether the influencing factor is a key influencing factor is determined based on the Pearson correlation coefficient.
8. The commodity price index prediction system based on multiple factors according to claim 7 is characterized in that: When the data processing module is used to determine whether the influencing factor is a key influencing factor according to the Pearson correlation coefficient, it includes: There is a preset Pearson correlation coefficient; Comparing the Pearson correlation coefficient with the preset Pearson correlation coefficient; If the Pearson correlation coefficient is greater than or equal to the preset Pearson correlation coefficient, determining that the influencing factor corresponding to the Pearson correlation coefficient is a key influencing factor; If the Pearson correlation coefficient is less than the preset Pearson correlation coefficient, it is determined that the influencing factor corresponding to the Pearson correlation coefficient is not a key influencing factor.
9. The commodity price index prediction system based on multiple factors according to claim 2, characterized in that: The model building module, when used to set the weights of the key influencing factors and generate a commodity price index model, includes: Establish a commodity price index forecasting model based on the key influencing factors; Training the commodity price index prediction model according to the influencing factor data to obtain multiple sets of coefficient values; Determining weights of the key influencing factors according to the multiple groups of coefficient values, fitting the weights with the commodity price index prediction model, and generating the commodity price index model; The expression of the commodity price index model is: K=αN D +βR D +γT D ; Among them, α represents the weight of the supply and demand of commodities, N D It is expressed as the supply and demand of commodities, β is expressed as the weight of the nonlinear coefficient of the production and sales location of commodities, R D It is expressed as the nonlinear coefficient of production place-sales place, γ is the weight of weather temperature when commodities are sold, T D Indicates how cold or warm the weather is when the commodity is sold.
10. The commodity price index prediction system based on multiple factors according to claim 1, characterized in that: When the price prediction module is used to predict the commodity price index according to the commodity price index model, it includes: Obtain data on key influencing factors; Based on the commodity price index model, the commodity price index is predicted according to the key influencing factor data to obtain a predicted value of the commodity price index.
Citation Information
Cited By
Product price index analysis method, system and device based on data processing
CN120013591A
