CN117635202B - Carbon asset price prediction method, device, electronic equipment and storage medium - Google Patents
Carbon asset price prediction method, device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN117635202B CN117635202B CN202410102230.2A CN202410102230A CN117635202B CN 117635202 B CN117635202 B CN 117635202B CN 202410102230 A CN202410102230 A CN 202410102230A CN 117635202 B CN117635202 B CN 117635202B
- Authority
- CN
- China
- Prior art keywords
- carbon asset
- scale
- prediction
- asset price
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 229910052799 carbon Inorganic materials 0.000 title claims abstract description 308
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 title claims abstract description 307
- 238000000034 method Methods 0.000 title claims abstract description 52
- 238000004519 manufacturing process Methods 0.000 claims description 92
- 238000012549 training Methods 0.000 claims description 16
- 238000004590 computer program Methods 0.000 claims description 13
- 230000015654 memory Effects 0.000 claims description 8
- 230000003442 weekly effect Effects 0.000 claims description 7
- 230000006870 function Effects 0.000 description 12
- 238000012545 processing Methods 0.000 description 11
- 230000008569 process Effects 0.000 description 8
- 230000009471 action Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 238000004891 communication Methods 0.000 description 4
- 230000003287 optical effect Effects 0.000 description 3
- 238000013135 deep learning Methods 0.000 description 2
- 239000004065 semiconductor Substances 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0206—Price or cost determination based on market factors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Development Economics (AREA)
- Theoretical Computer Science (AREA)
- Strategic Management (AREA)
- Computational Mathematics (AREA)
- Entrepreneurship & Innovation (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Physics (AREA)
- Evolutionary Biology (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Algebra (AREA)
- Probability & Statistics with Applications (AREA)
- Operations Research (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Game Theory and Decision Science (AREA)
- Economics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application relates to a carbon asset price prediction method, a device, electronic equipment and a storage medium, which are applied to the technical field of carbon assets, wherein the method comprises the following steps: aiming at a single scale period of M scale periods, acquiring K carbon asset price data and GDP data of the scale periods before the current moment in a target area and carbon asset industry characteristic data of the scale periods; inputting the K GDP data into the regional GDP prediction model respectively to obtain corresponding macroscopic prediction results; splicing the characteristic data of the carbon asset industry, the price data of the carbon asset in the K scale periods and the K macroscopic prediction results to obtain splicing characteristics; inputting the spliced characteristics into a prediction model corresponding to the scale period to obtain a carbon asset price prediction result corresponding to the scale period; and carrying out weighted average on the carbon asset price prediction results corresponding to the K scale periods respectively to obtain the carbon asset price prediction result of the target scale period after the current moment. The accuracy of the prediction can be improved.
Description
Technical Field
The present disclosure relates to the field of carbon asset technologies, and in particular, to a method and apparatus for predicting price of a carbon asset, an electronic device, and a storage medium.
Background
Currently, carbon asset price prediction is becoming increasingly important as quota carbon asset trial is pursued. In the related art, the prediction result may be obtained through learning of the history data based on a deep learning method. For example, based on an LSTM (Long Short-Term Memory) model, a predicted result can be obtained. The method predicts a relatively fixed period, for example, typically in days or weeks. The carbon asset price data sequences have different meanings on different scale periods and are closely related with the economic environment, so that the prediction failure phenomenon easily occurs, namely, the subsequent prediction results completely deviate from the true values from a certain time node. In addition, the Euclidean distance loss is calculated simply by adopting the price of the carbon asset in the loss function processing, so that the feasible solution is excessive, and model training is often not converged to a better feasible solution. It can be seen that the accuracy of predicting the price of a carbon asset based on the deep learning method is low.
Disclosure of Invention
In order to solve the technical problems, the application provides a carbon asset price prediction method, a device, electronic equipment and a storage medium.
According to a first aspect of the present application, there is provided a method of predicting price of a carbon asset, comprising:
Aiming at a single scale period of preset M different scale periods, acquiring carbon asset price data and domestic total production value data of K scale periods before the current moment in a target area, and carbon asset industry characteristic data corresponding to the scale periods; m is an integer greater than 1, K is a positive integer;
respectively inputting the domestic production total value data of the K scale periods into a regional domestic production total value prediction model corresponding to the scale periods to obtain macroscopic prediction results respectively corresponding to the domestic production total value data of the K scale periods;
splicing the carbon asset industry characteristic data corresponding to the scale period, the carbon asset price data of the K scale periods and the K macro prediction results to obtain splicing characteristics of the scale periods;
inputting the spliced characteristics of the scale period into a pre-trained prediction model corresponding to the scale period to obtain a carbon asset price prediction result corresponding to the scale period;
and carrying out weighted average on the carbon asset price prediction results corresponding to the M scale periods respectively to obtain the carbon asset price prediction results of the target scale period after the current moment, wherein the target scale period is the minimum scale period of the M scale periods.
Optionally, for each scale period, the training method of the prediction model corresponding to the scale period includes:
acquiring carbon asset price data and domestic production total value data in a historical time period;
dividing the carbon asset price data and the domestic production total value data in the historical time period according to the scale period to obtain a plurality of groups of carbon asset price data sequences and a plurality of groups of domestic production total value data sequences, wherein each group of carbon asset price data sequences comprises K sample carbon asset price data, and each group of domestic production total value data sequences comprises K sample domestic production total value data;
determining an actual prediction result corresponding to each group of carbon asset price data sequence from the carbon asset price data in the historical time period;
inputting each sample domestic production total value data in each group of domestic production total value data sequences into a regional domestic production total value prediction model corresponding to the scale period to obtain a sample macroscopic prediction result;
splicing the carbon asset industry characteristic data, the K sample carbon asset price data and the K sample macroscopic prediction results corresponding to the scale period to obtain sample splicing characteristics of the scale period;
Inputting the sample splicing characteristics into an initial model to obtain a model prediction result corresponding to the carbon asset price data sequence;
performing 1-n-step gradient operation on actual prediction results corresponding to the multi-group carbon asset price data sequences to obtain 1-n-step actual gradient results; n is a positive integer;
performing 1-n-step-degree operation on model prediction results corresponding to the multi-group carbon asset price data sequences to obtain 1-n-step prediction gradient results;
performing a second-norm operation on the i-th order actual gradient result and the i-th order predicted gradient result to obtain an i-th order norm operation result; wherein i is an integer of 0~n, the 0 th order actual gradient result is the actual prediction result, and the 0 th order prediction gradient result is the model prediction result;
carrying out weighted average on the 0~n th order norm operation result to obtain a loss value;
and updating the network parameters in the initial model through the loss value until the updated model converges, and determining the finally updated model as a prediction model corresponding to the scale period.
Optionally, the M scale periods are specifically 3 scale periods, where the 3 scale periods include: week, month and year;
the weighted average of the carbon asset price prediction results corresponding to the M scale periods respectively is carried out to obtain the carbon asset price prediction result of the target scale period after the current moment, and the method comprises the following steps:
If the weekly corresponding carbon asset price prediction result is cweek, the monthly corresponding carbon asset price prediction result is cmonth, the annual corresponding carbon asset price prediction result is cyear, and the formula is as follows:determining a carbon asset price prediction result C of one week after the current moment; wherein (1)>Is a weight parameter;
。
optionally, the splicing the carbon asset industry characteristic data corresponding to the scale period, the K carbon asset price data of the scale period and the K macro prediction results to obtain the splicing characteristic of the scale period includes:
for each carbon asset price data in the K carbon asset price data of the scale period, splicing the carbon asset industry characteristic data with the carbon asset price data and a macroscopic prediction result corresponding to the carbon asset price data to obtain splicing characteristics corresponding to the carbon asset price data;
and splicing the splicing characteristics corresponding to the K carbon asset price data to obtain the splicing characteristics of the scale period.
Optionally, the carbon asset industry characteristic data corresponding to the scale period includes: industry profit margin, industry productivity and industry stock ratio features.
According to a second aspect of the present application, there is provided a carbon asset price prediction device comprising:
The data acquisition module is used for acquiring carbon asset price data and domestic total production value data of K scale periods before the current moment in a target area and carbon asset industry characteristic data corresponding to the scale periods according to preset single scale periods in M different scale periods; m is an integer greater than 1, K is a positive integer;
the first prediction module is used for respectively inputting the domestic production total value data of the K scale periods into the regional domestic production total value prediction model corresponding to the scale periods to obtain macroscopic prediction results respectively corresponding to the domestic production total value data of the K scale periods;
the prediction result splicing module is used for splicing the carbon asset industry characteristic data corresponding to the scale period, the carbon asset price data of the K scale periods and the K macroscopic prediction results to obtain splicing characteristics of the scale periods;
the second prediction module is used for inputting the spliced characteristics of the scale period into a pre-trained prediction model corresponding to the scale period to obtain a carbon asset price prediction result corresponding to the scale period;
and the final prediction result determining module is used for carrying out weighted average on the carbon asset price prediction results respectively corresponding to the M scale periods to obtain the carbon asset price prediction result of the target scale period after the current moment, wherein the target scale period is the minimum scale period of the M scale periods.
Optionally, the apparatus further comprises:
the prediction model training module is used for training the prediction model corresponding to each scale period through the following steps:
acquiring carbon asset price data and domestic production total value data in a historical time period;
dividing the carbon asset price data and the domestic production total value data in the historical time period according to the scale period to obtain a plurality of groups of carbon asset price data sequences and a plurality of groups of domestic production total value data sequences, wherein each group of carbon asset price data sequences comprises K sample carbon asset price data, and each group of domestic production total value data sequences comprises K sample domestic production total value data;
determining an actual prediction result corresponding to each group of carbon asset price data sequence from the carbon asset price data in the historical time period;
inputting each sample domestic production total value data in each group of domestic production total value data sequences into a regional domestic production total value prediction model corresponding to the scale period to obtain a sample macroscopic prediction result;
splicing the carbon asset industry characteristic data, the K sample carbon asset price data and the K sample macroscopic prediction results corresponding to the scale period to obtain sample splicing characteristics of the scale period;
Inputting the sample splicing characteristics into an initial model to obtain a model prediction result corresponding to the carbon asset price data sequence;
performing 1-n-step gradient operation on actual prediction results corresponding to the multi-group carbon asset price data sequences to obtain 1-n-step actual gradient results; n is a positive integer;
performing 1-n-step-degree operation on model prediction results corresponding to the multi-group carbon asset price data sequences to obtain 1-n-step prediction gradient results;
performing a second-norm operation on the i-th order actual gradient result and the i-th order predicted gradient result to obtain an i-th order norm operation result; wherein i is an integer of 0~n, the 0 th order actual gradient result is the actual prediction result, and the 0 th order prediction gradient result is the model prediction result;
carrying out weighted average on the 0~n th order norm operation result to obtain a loss value;
and updating the network parameters in the initial model through the loss value until the updated model converges, and determining the finally updated model as a prediction model corresponding to the scale period.
Optionally, the M scale periods are specifically 3 scale periods, where the 3 scale periods include: week, month and year;
the final prediction result determining module is specifically configured to, if the weekly corresponding carbon asset price prediction result is cweek, the monthly corresponding carbon asset price prediction result is cmonth, the annual corresponding carbon asset price prediction result is cyearr, according to the formula:
Determining a carbon asset price prediction result C of one week after the current moment; wherein (1)>Is a weight parameter;
。
optionally, the prediction result splicing module is specifically configured to splice, for each piece of carbon asset price data in the K scale periods, the carbon asset industry feature data with the carbon asset price data and a macroscopic prediction result corresponding to the carbon asset price data, so as to obtain a splice feature corresponding to the carbon asset price data; and splicing the splicing characteristics corresponding to the K carbon asset price data to obtain the splicing characteristics of the scale period.
Optionally, the carbon asset industry characteristic data corresponding to the scale period includes: industry profit margin, industry productivity and industry stock ratio features.
According to a third aspect of the present application, there is provided an electronic device comprising: a processor for executing a computer program stored in a memory, which when executed by the processor implements the method according to the first aspect.
According to a fourth aspect of the present application, there is provided a computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the method of the first aspect.
According to a fifth aspect of the present application, there is provided a computer program product for, when run on a computer, causing the computer to perform the method of the first aspect.
Compared with the prior art, the technical scheme provided by the embodiment of the application has the following advantages:
predicting the price of the carbon asset from the dimension of a single scale period by presetting a plurality of scale periods, and carrying out weighted average on the carbon asset price prediction results corresponding to the scale periods respectively to obtain a final prediction result. In this way, by predicting the price of the carbon asset from the dimensions of the multiple scale periods, the meaning of the carbon asset price data sequence on different scale periods can be extracted, thereby improving the accuracy of the price prediction of the carbon asset. In the process of predicting the price of the carbon asset from the dimension of a single scale period, processing domestic production total value data through a regional domestic production total value prediction model to obtain a macroscopic prediction result, and then splicing historical carbon asset price sequence data with the macroscopic prediction result and carbon asset industry characteristic data to obtain splicing characteristics. And processing the splicing characteristics through a prediction model to obtain a prediction result. The accuracy of carbon asset price prediction can be improved by combining a macroscopic economic prediction model (i.e., a regional domestic total production value prediction model), a microscopic economic prediction model (carbon asset industry characteristic data related to supply-demand relationships), and carbon asset industry characteristic data closely related to carbon asset prices.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the application and together with the description, serve to explain the principles of the application.
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings that are required to be used in the description of the embodiments or the prior art will be briefly described below, and it will be obvious to those skilled in the art that other drawings can be obtained from these drawings without inventive effort.
FIG. 1 is a flow chart of a method of predicting price of a carbon asset in an embodiment of the present application;
FIG. 2 is a schematic diagram of a network structure for price prediction of carbon assets in an embodiment of the present application;
FIG. 3 is a flowchart of a predictive model training method in an embodiment of the present application;
FIG. 4 is a schematic diagram of a carbon asset price prediction device according to an embodiment of the present application;
fig. 5 is a schematic structural diagram of an electronic device in an embodiment of the present application.
Detailed Description
In order that the above objects, features and advantages of the present application may be more clearly understood, a further description of the aspects of the present application will be provided below. It should be noted that, in the case of no conflict, the embodiments of the present application and the features in the embodiments may be combined with each other.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present application, but the present application may be practiced otherwise than as described herein; it will be apparent that the embodiments in the specification are only some, but not all, embodiments of the application.
Referring to fig. 1, fig. 1 is a flowchart of a method for predicting price of a carbon asset according to an embodiment of the present application, which may include the following steps:
step S102, aiming at a single scale period in M preset different scale periods, acquiring K carbon asset price data and domestic total production value data of the scale periods before the current moment in a target area and carbon asset industry characteristic data corresponding to the scale periods.
Scale period refers to the period used to predict the price of a carbon asset and may be weeks, months, quarters, years, etc. In this embodiment of the present application, M scale periods may be preset, where M is an integer greater than 1, for example, M may be 2, 3, 4, etc., that is, a plurality of scale periods are preset. Optionally, the M scale periods are specifically 3 scale periods, where the 3 scale periods include: week, month and year.
The process of predicting price of a carbon asset in the dimensions of different scale periods is similar in M scale periods, a single scale period being illustrated herein as an example. For a single scale period, the prediction of the carbon asset price can be performed according to historical carbon asset price data, and the closer the historical carbon asset price data is to the current moment, the higher the accuracy of the predicted carbon asset price. Thus, K carbon asset price data for the scale period may be obtained for the target region prior to the current time, where K is a positive integer. For example, the scale period is a week, carbon asset price data for K weeks before the current time may be acquired, alternatively, the carbon asset price data for K weeks may be carbon asset price data for K consecutive weeks. For example, the scale period is month, and carbon asset price data for K months prior to the current time may be obtained. And then, predicting the price of the carbon asset according to the price data of the carbon asset in K scale periods.
The characteristic data of the carbon asset industry is closely related to the price of the carbon asset, even the price of the carbon asset is determined, and the characteristic data plays an important characteristic expression role in predicting the price of the carbon asset. Therefore, the characteristic data of the carbon asset industry corresponding to each scale period can be obtained, and the price of the carbon asset is predicted by combining the characteristic data of the carbon asset industry, so that the accuracy of the price prediction of the carbon asset is improved. For example, the 3 scale periods include: and the week, month and year can obtain the characteristic data of the carbon asset industry for predicting the price of the carbon asset on the scale of the week, month and year.
Optionally, the carbon asset industry characteristic data corresponding to the scale period includes: industry profit margin, industry productivity and industry stock ratio features. Industry profit margin refers to the average industry profit margin of the industry or industries in which the predicted carbon asset price is located, a feature that is primarily representative of the ability and willingness of the carbon asset to buy and sell. Industry capacity refers to the average industry capacity of the industry or industries in which the predicted price of the carbon asset is located, and this characteristic represents the consumption capacity of the carbon asset. The industry stock ratio feature refers to the ratio of the industry residual carbon asset of the industry or the average industry of the plurality of industries in which the predicted carbon asset price is located to the starting quota total asset, and represents the scarcity and consumption degree of the carbon asset.
Step S104, inputting the domestic production total value data of the K scale periods into the regional domestic production total value prediction model corresponding to the scale periods respectively, and obtaining macroscopic prediction results corresponding to the domestic production total value data of the K scale periods respectively.
The regional domestic total production value prediction model corresponding to each scale period is a pre-trained offline model, belongs to a macroscopic economic prediction model, and is mainly related to a large economic environment. Alternatively, the domestic total production value prediction model of the region corresponding to each scale period can be an LSTM model.
And inputting each domestic production total value data in the domestic production total value data of the K scale periods into a domestic production total value prediction model of a region corresponding to the scale period, so that a macroscopic prediction result corresponding to the domestic production total value data can be obtained, and K macroscopic prediction results are obtained.
And S106, splicing the carbon asset industry characteristic data corresponding to the scale period, the K carbon asset price data of the scale period and the K macroscopic prediction results to obtain the splicing characteristic of the scale period.
The number of the carbon asset price data and the macroscopic prediction result is K, the number of the carbon asset industry characteristic data is 1, and the 1 carbon asset industry characteristic data can comprise a plurality of contents, such as the industry profit margin, the industry productivity, the industry stock ratio characteristic and the like.
Alternatively, the stitching of the data may be performed as follows:
and splicing the carbon asset industry characteristic data with the macroscopic prediction result corresponding to the carbon asset price data aiming at each carbon asset price data in the K carbon asset price data of the scale period to obtain the splicing characteristic corresponding to the carbon asset price data. And then, splicing the splicing characteristics corresponding to the price data of the K carbon assets to obtain the splicing characteristics of the scale period.
Specifically, the carbon asset industry characteristic data is denoted as V1, the K carbon asset price data for the scale cycle is denoted as V21, V22, …, V2K, and the K macro predictions are denoted as V31, V32, …, V3K. The characteristic vector v11, v12, … and v1K can be formed by repeating the characteristic data of the carbon asset industry for K times; where v11=v12=v1k. The splice characteristics of the scale period after splicing can be expressed as [ v21, v31, v11], [ v22, v32, v12], …, [ v2kv3k, v1k ]. It can be seen that the splicing characteristic is obtained by splicing the three-dimensional data to obtain three-dimensional characteristics and then splicing the K three-dimensional characteristics, so that the accuracy of predicting the price of the carbon asset can be improved by utilizing the sequence characteristics in the splicing characteristics.
Step S108, inputting the spliced characteristics of the scale period into a pre-trained prediction model corresponding to the scale period to obtain a carbon asset price prediction result corresponding to the scale period.
The prediction model of the step corresponds to the macroscopic economic prediction model, belongs to the microscopic economic prediction model and is directly related to the price. Similar to the macro-economic prediction model, the prediction model may also be an LSTM model, and a method for training the prediction model will be described below.
And step S110, carrying out weighted average on the carbon asset price prediction results corresponding to the M scale periods respectively to obtain the carbon asset price prediction result of the target scale period after the current moment.
In the embodiment of the application, the target scale period may be the minimum scale period of the M scale periods, that is, the carbon asset price prediction result of the minimum scale period after the current time is predicted, so that the accuracy of the predicted carbon asset price is higher. For example, the M scale periods are specifically 3 scale periods, namely week, month and year, respectively, and the price of the carbon asset in the week next to the current moment can be predicted.
It will be appreciated that the accuracy of the carbon asset price predictions for different scale periods will be different, and therefore the weights of the carbon asset price predictions for different scale periods will be different. The smaller the scale period, the more accurate the carbon asset price prediction results, and the greater the corresponding weights.
Alternatively, the M scale periods are specifically 3 scale periods, respectively weeks, months and years. If the weekly carbon asset price prediction result is cweek, the monthly carbon asset price prediction result is cmonth, and the annual carbon asset price prediction result is cyear, the formula may be:determining a carbon asset price prediction result C of one week after the current moment; wherein (1)>Is a weight parameter;
since a year comprises 12 months, a month comprises 4 weeks,specifically, the method can be calculated by the following formula:
。
the weight parameters are calculated according to the formula, and the price of the carbon asset is predicted according to the weight parameters, so that the accuracy of the price prediction of the carbon asset can be improved.
Referring to fig. 2, fig. 2 is a schematic diagram of a network structure for predicting price of a carbon asset according to an embodiment of the present application, which includes three scale periods, and each scale period is processed similarly. And respectively obtaining the carbon asset price prediction results corresponding to the three scale periods, and then carrying out weighted average on the three carbon asset price prediction results to obtain a final carbon asset price prediction result.
According to the carbon asset price prediction method, the carbon asset price is predicted from the dimension of a single scale period by presetting a plurality of scale periods, and the carbon asset price prediction results corresponding to the scale periods are weighted and averaged to obtain a final prediction result. In this way, by predicting the price of the carbon asset from the dimensions of the multiple scale periods, the meaning of the carbon asset price data sequence on different scale periods can be extracted, thereby improving the accuracy of the price prediction of the carbon asset. In the process of predicting the price of the carbon asset from the dimension of a single scale period, processing domestic production total value data through a regional domestic production total value prediction model to obtain a macroscopic prediction result, and then splicing historical carbon asset price sequence data with the macroscopic prediction result and carbon asset industry characteristic data to obtain splicing characteristics. And processing the splicing characteristics through a prediction model to obtain a prediction result. The accuracy of carbon asset price prediction can be improved by combining a macroscopic economic prediction model (i.e., a regional domestic total production value prediction model), a microscopic economic prediction model (carbon asset industry characteristic data related to supply-demand relationships), and carbon asset industry characteristic data closely related to carbon asset prices.
In the embodiment of the present application, the training method of the prediction model corresponding to different scale periods is similar, and the training method of the prediction model corresponding to one scale period is described here as an example.
Referring to fig. 3, fig. 3 is a flowchart of a method for training a prediction model according to an embodiment of the present application, which may include the following steps:
step S302, acquiring carbon asset price data and domestic production total value data in a historical time period.
The historical time period can be the historical time period which is the latest with the current moment, and the length of the selected historical time period can be different according to different scale periods. For example, if the scale period is weekly, the historical time period may be 1 year prior to the current time; if the scale period is a month, the historical time period may be 3 years before the current time, etc.; if the scale period is a year, the historical time period may be 10 years before the current time, or the like.
And step S304, respectively dividing the carbon asset price data and the domestic production total value data in the historical time period according to the scale period to obtain a plurality of groups of carbon asset price data sequences and a plurality of groups of domestic production total value data sequences. The carbon asset price data sequence of each group comprises K sample carbon asset price data, and the domestic production total value data sequence of each group comprises K sample domestic production total value data.
In the embodiment of the application, the carbon asset price data and the domestic production total value data in the historical time period can be respectively divided according to the scale period to obtain a plurality of continuous sample carbon asset price data and sample domestic production total value data of the scale period. K consecutive sample carbon asset price data may be used as a set of carbon asset price data sequences and K consecutive sample domestic production total data may be used as a set of domestic production total data sequences. The carbon asset price data sequence and the domestic production total data sequence are corresponding in the time dimension.
For example, the scale period is week, the value of K is 3, the history period is one year before the current time, and the history period can be divided into 52 weeks, so as to obtain carbon asset price data and domestic production total value data corresponding to the 52 weeks respectively. The continuous 3 carbon asset price data are taken as a group to form a group of carbon asset price data sequence, and the continuous 3 domestic production total value data are taken as a group to form a group of domestic production total value data sequence.
And step S306, determining the actual prediction result corresponding to each group of carbon asset price data sequence from the carbon asset price data in the historical time period.
For each group of carbon asset price data sequences, carbon asset price data of the next scale period corresponding to the group of carbon asset price data sequences can be determined, and the carbon asset price data of the next scale period is used as an actual prediction result corresponding to the group of carbon asset price data sequences.
For example, the current carbon asset price data sequence is the current 3 carbon asset price data, and then the corresponding actual prediction result is the carbon asset price data following the current 3 carbon asset price data.
And step 308, inputting the domestic production total value data of each sample in each group of domestic production total value data sequences into a regional domestic production total value prediction model corresponding to the scale period to obtain a sample macroscopic prediction result.
And step S310, splicing the carbon asset industry characteristic data, the K sample carbon asset price data and the K sample macroscopic prediction results corresponding to the scale period to obtain sample splicing characteristics of the scale period.
The processing procedure of step S308 to step S310 is the same as the processing procedure of step S104 to step S106 in the embodiment of fig. 1, and may be specifically referred to the description in the embodiment of fig. 1, and will not be repeated here.
And step S312, inputting the sample splicing characteristics into an initial model to obtain a model prediction result corresponding to the carbon asset price data sequence.
The initial model has the same model structure as the trained prediction model, the network parameter values are different, and after training, the network parameter values are continuously adjusted, and the initial model is updated into the prediction model.
Step S314, carrying out 1-n-step gradient operation on the actual prediction results corresponding to the multi-carbon asset price data sequences to obtain 1-n-step actual gradient results; and carrying out 1-n-step-degree operation on model prediction results corresponding to the multi-group carbon asset price data sequences to obtain 1-n-step prediction gradient results. Wherein n is a positive integer.
After the actual prediction result and the model prediction result are obtained, substituting the actual prediction result and the model prediction result into a pre-constructed loss function to obtain a loss value. In the embodiment of the application, the multi-order gradient loss function can be constructed, and the multi-order gradient loss function can apply more constraint to model training, so that the model is more accurate in expressing the high-gradient characteristics of data.
Step S316, performing a second-norm operation on the i-th order actual gradient result and the i-th order predicted gradient result to obtain an i-th order norm operation result, and performing a weighted average on the 0~n-th order norm operation result to obtain a loss value, wherein i is an integer of 0~n.
According to the loss function formula:a loss value loss is obtained.
The 0 th order actual gradient result is an actual prediction result, and the 0 th order prediction gradient result is a model prediction result.GiRepresenting the result of the actual gradient of the i-th order,girepresenting the result of the i-th order prediction gradient. It can be seen that when n is 0, the above-mentioned loss function is degraded to an existing euclidean distance loss function.
And step S318, updating the network parameters in the initial model through the loss value until the updated model converges, and determining the finally updated model as a prediction model corresponding to the scale period.
According to the method and the device, the model can be trained under the multi-order gradient by constructing the multi-order gradient loss function for predicting the carbon asset price sequence, so that accuracy of sequence prediction can be maintained.
The embodiment of the application also provides a carbon asset price prediction device, referring to fig. 4, the carbon asset price prediction device 400 includes:
the data acquisition module 402 is configured to acquire, for a single scale period of M preset different scale periods, K carbon asset price data and domestic total production value data of the scale periods before a current time in a target area, and carbon asset industry characteristic data corresponding to the scale periods; m is an integer greater than 1, K is a positive integer;
The first prediction module 404 is configured to input the domestic total production value data of the K scale periods into the regional domestic total production value prediction model corresponding to the scale periods, so as to obtain macroscopic prediction results corresponding to the domestic total production value data of the K scale periods;
the prediction result splicing module 406 is configured to splice the carbon asset industry feature data corresponding to the scale period, the carbon asset price data of the K scale periods, and the K macro prediction results, to obtain a splice feature of the scale period;
a second prediction module 408, configured to input the spliced feature of the scale period into a pre-trained prediction model corresponding to the scale period, to obtain a carbon asset price prediction result corresponding to the scale period;
the final prediction result determining module 410 is configured to perform weighted average on the carbon asset price prediction results corresponding to the M scale periods respectively, so as to obtain a carbon asset price prediction result of a target scale period after the current moment, where the target scale period is a minimum scale period of the M scale periods.
Optionally, the carbon asset price prediction apparatus 400 further includes:
the prediction model training module is used for training the prediction model corresponding to each scale period through the following steps:
Acquiring carbon asset price data and domestic production total value data in a historical time period;
dividing the carbon asset price data and the domestic production total value data in the historical time period according to the scale period to obtain a plurality of groups of carbon asset price data sequences and a plurality of groups of domestic production total value data sequences, wherein each group of carbon asset price data sequences comprises K sample carbon asset price data, and each group of domestic production total value data sequences comprises K sample domestic production total value data;
determining an actual prediction result corresponding to each group of carbon asset price data sequence from the carbon asset price data in the historical time period;
inputting the domestic production total value data of each sample in each group of domestic production total value data sequences into a regional domestic production total value prediction model corresponding to the scale period to obtain a sample macroscopic prediction result;
splicing the carbon asset industry characteristic data corresponding to the scale period, the K sample carbon asset price data and the K sample macroscopic prediction results to obtain sample splicing characteristics of the scale period;
inputting the sample splicing characteristics into an initial model to obtain a model prediction result corresponding to the carbon asset price data sequence;
performing 1-n-step gradient operation on actual prediction results corresponding to the multi-group carbon asset price data sequences to obtain 1-n-step actual gradient results; n is a positive integer;
Performing 1-n-step-degree operation on model prediction results corresponding to the multi-group carbon asset price data sequences to obtain 1-n-step prediction gradient results;
performing a second-norm operation on the i-th order actual gradient result and the i-th order predicted gradient result to obtain an i-th order norm operation result; wherein i is an integer of 0~n, the 0 th order actual gradient result is an actual prediction result, and the 0 th order prediction gradient result is a model prediction result;
carrying out weighted average on the 0~n th order norm operation result to obtain a loss value;
and updating network parameters in the initial model through the loss value until the updated model converges, and determining the finally updated model as a prediction model corresponding to the scale period.
Optionally, the M scale periods are specifically 3 scale periods, where the 3 scale periods include: week, month and year;
the final prediction result determining module 410 is specifically configured to, if the weekly corresponding carbon asset price prediction result is cweek, the monthly corresponding carbon asset price prediction result is cmonth, the annual corresponding carbon asset price prediction result is cyearr, according to the formula:
determining a carbon asset price prediction result C of one week after the current moment; wherein (1)>Is a weight parameter;
。
optionally, the prediction result splicing module 406 is specifically configured to splice, for each piece of carbon asset price data in the K scale periods, the carbon asset industry feature data with the carbon asset price data and a macroscopic prediction result corresponding to the carbon asset price data, so as to obtain a splice feature corresponding to the carbon asset price data; and splicing the splicing characteristics corresponding to the K carbon asset price data to obtain the splicing characteristics of the scale period.
Optionally, the carbon asset industry characteristic data corresponding to the scale period includes: industry profit margin, industry productivity and industry stock ratio features.
Specific details of each module or unit in the above apparatus have been described in the corresponding method, and thus are not described herein.
It should be noted that although in the above detailed description several modules or units of a device for action execution are mentioned, such a division is not mandatory. Indeed, the features and functions of two or more modules or units described above may be embodied in one module or unit, in accordance with embodiments of the present application. Conversely, the features and functions of one module or unit described above may be further divided into a plurality of modules or units to be embodied.
In an exemplary embodiment of the present application, there is also provided an electronic apparatus including: a processor; a memory for storing processor-executable instructions; wherein the processor is configured to perform the carbon asset price prediction method described above in this example embodiment.
Fig. 5 is a schematic structural diagram of an electronic device in an embodiment of the present application. It should be noted that, the electronic device 500 shown in fig. 5 is only an example, and should not impose any limitation on the functions and the application scope of the embodiments of the present application.
As shown in fig. 5, the electronic device 500 includes a Central Processing Unit (CPU) 501, which can perform various appropriate actions and processes according to a program stored in a Read Only Memory (ROM) 502 or a program loaded from a storage section 508 into a Random Access Memory (RAM) 503. In the RAM 503, various programs and data required for the system operation are also stored. The central processing unit 501, the ROM 502, and the RAM 503 are connected to each other via a bus 504. An input/output (I/O) interface 505 is also connected to bus 504.
The following components are connected to the I/O interface 505: an input section 506 including a keyboard, a mouse, and the like; an output portion 507 including a Cathode Ray Tube (CRT), a Liquid Crystal Display (LCD), and the like, and a speaker, and the like; a storage portion 508 including a hard disk and the like; and a communication section 509 including a network interface card such as a Local Area Network (LAN) card, a modem, or the like. The communication section 509 performs communication processing via a network such as the internet. The drive 510 is also connected to the I/O interface 505 as needed. A removable medium 511 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like is mounted on the drive 510 as needed so that a computer program read therefrom is mounted into the storage section 508 as needed.
In particular, according to embodiments of the present application, the processes described above with reference to flowcharts may be implemented as computer software programs. For example, embodiments of the present application include a computer program product comprising a computer program embodied on a computer readable medium, the computer program comprising program code for performing the method shown in the flowcharts. In such an embodiment, the computer program may be downloaded and installed from a network via the communication portion 509, and/or installed from the removable media 511. When being executed by the central processing unit 501, performs the various functions defined in the apparatus of the present application.
In an embodiment of the present application, there is also provided a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the above-described carbon asset price prediction method.
The computer readable storage medium shown in the present application may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples of the computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory, a read-only memory, an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the context of this document, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable storage medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, fiber optic cable, radio frequency, and the like, or any suitable combination of the foregoing.
In an embodiment of the present application, there is also provided a computer program product, which when run on a computer causes the computer to perform the above-described carbon asset price prediction method.
It should be noted that in this document, relational terms such as "first" and "second" and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
The foregoing is merely a specific embodiment of the application to enable one skilled in the art to understand or practice the application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown and described herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (10)
1. A method of predicting price of a carbon asset, comprising:
aiming at a single scale period of preset M different scale periods, acquiring carbon asset price data and domestic total production value data of K scale periods before the current moment in a target area, and carbon asset industry characteristic data corresponding to the scale periods; m is an integer greater than 1, K is a positive integer;
respectively inputting the domestic production total value data of the K scale periods into a regional domestic production total value prediction model corresponding to the scale periods to obtain macroscopic prediction results respectively corresponding to the domestic production total value data of the K scale periods;
splicing the carbon asset industry characteristic data corresponding to the scale period, the carbon asset price data of the K scale periods and the K macro prediction results to obtain splicing characteristics of the scale periods;
inputting the spliced characteristics of the scale period into a pre-trained prediction model corresponding to the scale period to obtain a carbon asset price prediction result corresponding to the scale period;
and carrying out weighted average on the carbon asset price prediction results corresponding to the M scale periods respectively to obtain the carbon asset price prediction results of the target scale period after the current moment, wherein the target scale period is the minimum scale period of the M scale periods.
2. The method according to claim 1, wherein for each scale period, the training method of the prediction model corresponding to the scale period comprises:
acquiring carbon asset price data and domestic production total value data in a historical time period;
dividing the carbon asset price data and the domestic production total value data in the historical time period according to the scale period to obtain a plurality of groups of carbon asset price data sequences and a plurality of groups of domestic production total value data sequences, wherein each group of carbon asset price data sequences comprises K sample carbon asset price data, and each group of domestic production total value data sequences comprises K sample domestic production total value data;
determining an actual prediction result corresponding to each group of carbon asset price data sequence from the carbon asset price data in the historical time period;
inputting each sample domestic production total value data in each group of domestic production total value data sequences into a regional domestic production total value prediction model corresponding to the scale period to obtain a sample macroscopic prediction result;
splicing the carbon asset industry characteristic data, the K sample carbon asset price data and the K sample macroscopic prediction results corresponding to the scale period to obtain sample splicing characteristics of the scale period;
Inputting the sample splicing characteristics into an initial model to obtain a model prediction result corresponding to the carbon asset price data sequence;
performing 1-n-step gradient operation on actual prediction results corresponding to the multi-group carbon asset price data sequences to obtain 1-n-step actual gradient results; n is a positive integer;
performing 1-n-step-degree operation on model prediction results corresponding to the multi-group carbon asset price data sequences to obtain 1-n-step prediction gradient results;
performing a second-norm operation on the i-th order actual gradient result and the i-th order predicted gradient result to obtain an i-th order norm operation result; wherein i is an integer of 0~n, the 0 th order actual gradient result is the actual prediction result, and the 0 th order prediction gradient result is the model prediction result;
carrying out weighted average on the 0~n th order norm operation result to obtain a loss value;
and updating the network parameters in the initial model through the loss value until the updated model converges, and determining the finally updated model as a prediction model corresponding to the scale period.
3. The method according to claim 1, wherein the M scale periods are in particular 3 scale periods, the 3 scale periods comprising: week, month and year;
The weighted average of the carbon asset price prediction results corresponding to the M scale periods respectively is carried out to obtain the carbon asset price prediction result of the target scale period after the current moment, and the method comprises the following steps:
if the weekly corresponding carbon asset price prediction result is cweek, the monthly corresponding carbon asset price prediction result is cmonth, the annual corresponding carbon asset price prediction result is cyear, and the formula is as follows:determining a carbon asset price prediction result C of one week after the current moment; wherein (1)>Is a weight parameter;
。
4. the method of claim 1, wherein the splicing the carbon asset industry feature data corresponding to the scale period, the K carbon asset price data for the scale period, and the K macro prediction results to obtain the spliced feature for the scale period comprises:
for each carbon asset price data in the K carbon asset price data of the scale period, splicing the carbon asset industry characteristic data with the carbon asset price data and a macroscopic prediction result corresponding to the carbon asset price data to obtain splicing characteristics corresponding to the carbon asset price data;
and splicing the splicing characteristics corresponding to the K carbon asset price data to obtain the splicing characteristics of the scale period.
5. The method of claim 1, wherein the carbon asset industry characteristic data corresponding to the scale period comprises: industry profit margin, industry productivity and industry stock ratio features.
6. A carbon asset price prediction device, the device comprising:
the data acquisition module is used for acquiring carbon asset price data and domestic total production value data of K scale periods before the current moment in a target area and carbon asset industry characteristic data corresponding to the scale periods according to preset single scale periods in M different scale periods; m is an integer greater than 1, K is a positive integer;
the first prediction module is used for respectively inputting the domestic production total value data of the K scale periods into the regional domestic production total value prediction model corresponding to the scale periods to obtain macroscopic prediction results respectively corresponding to the domestic production total value data of the K scale periods;
the prediction result splicing module is used for splicing the carbon asset industry characteristic data corresponding to the scale period, the carbon asset price data of the K scale periods and the K macroscopic prediction results to obtain splicing characteristics of the scale periods;
The second prediction module is used for inputting the spliced characteristics of the scale period into a pre-trained prediction model corresponding to the scale period to obtain a carbon asset price prediction result corresponding to the scale period;
and the final prediction result determining module is used for carrying out weighted average on the carbon asset price prediction results respectively corresponding to the M scale periods to obtain the carbon asset price prediction result of the target scale period after the current moment, wherein the target scale period is the minimum scale period of the M scale periods.
7. The apparatus of claim 6, wherein the apparatus further comprises:
the prediction model training module is used for training the prediction model corresponding to each scale period through the following steps:
acquiring carbon asset price data and domestic production total value data in a historical time period;
dividing the carbon asset price data and the domestic production total value data in the historical time period according to the scale period to obtain a plurality of groups of carbon asset price data sequences and a plurality of groups of domestic production total value data sequences, wherein each group of carbon asset price data sequences comprises K sample carbon asset price data, and each group of domestic production total value data sequences comprises K sample domestic production total value data;
Determining an actual prediction result corresponding to each group of carbon asset price data sequence from the carbon asset price data in the historical time period;
inputting each sample domestic production total value data in each group of domestic production total value data sequences into a regional domestic production total value prediction model corresponding to the scale period to obtain a sample macroscopic prediction result;
splicing the carbon asset industry characteristic data, the K sample carbon asset price data and the K sample macroscopic prediction results corresponding to the scale period to obtain sample splicing characteristics of the scale period;
inputting the sample splicing characteristics into an initial model to obtain a model prediction result corresponding to the carbon asset price data sequence;
performing 1-n-step gradient operation on actual prediction results corresponding to the multi-group carbon asset price data sequences to obtain 1-n-step actual gradient results; n is a positive integer;
performing 1-n-step-degree operation on model prediction results corresponding to the multi-group carbon asset price data sequences to obtain 1-n-step prediction gradient results;
performing a second-norm operation on the i-th order actual gradient result and the i-th order predicted gradient result to obtain an i-th order norm operation result; wherein i is an integer of 0~n, the 0 th order actual gradient result is the actual prediction result, and the 0 th order prediction gradient result is the model prediction result;
Carrying out weighted average on the 0~n th order norm operation result to obtain a loss value;
and updating the network parameters in the initial model through the loss value until the updated model converges, and determining the finally updated model as a prediction model corresponding to the scale period.
8. The apparatus of claim 6, wherein the M scale periods are specifically 3 scale periods, the 3 scale periods comprising: week, month and year;
the final prediction result determining module is specifically configured to, if the weekly corresponding carbon asset price prediction result is cweek, the monthly corresponding carbon asset price prediction result is cmonth, the annual corresponding carbon asset price prediction result is cyearr, according to the formula:determining a carbon asset price prediction result C of one week after the current moment; wherein (1)>Is a weight parameter;
。
9. an electronic device, comprising: a processor for executing a computer program stored in a memory, which when executed by the processor implements the method of any of claims 1-5.
10. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the method of any of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410102230.2A CN117635202B (en) | 2024-01-25 | 2024-01-25 | Carbon asset price prediction method, device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410102230.2A CN117635202B (en) | 2024-01-25 | 2024-01-25 | Carbon asset price prediction method, device, electronic equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117635202A CN117635202A (en) | 2024-03-01 |
| CN117635202B true CN117635202B (en) | 2024-04-12 |
Family
ID=90025473
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410102230.2A Active CN117635202B (en) | 2024-01-25 | 2024-01-25 | Carbon asset price prediction method, device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117635202B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105160204A (en) * | 2015-10-28 | 2015-12-16 | 河海大学 | Carbon emission price combination prediction method |
| CN115186507A (en) * | 2022-08-01 | 2022-10-14 | 广东电网有限责任公司电力调度控制中心 | Carbon value prediction method and device based on system dynamics |
| CN116452240A (en) * | 2023-04-13 | 2023-07-18 | 国网北京市电力公司 | Carbon transaction price prediction system, method, equipment and storage medium |
| WO2023146525A1 (en) * | 2022-01-27 | 2023-08-03 | Vincent Dert | Determining and/or evaluating a sustainability of a product, a service, an organization and/or a person |
-
2024
- 2024-01-25 CN CN202410102230.2A patent/CN117635202B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105160204A (en) * | 2015-10-28 | 2015-12-16 | 河海大学 | Carbon emission price combination prediction method |
| WO2023146525A1 (en) * | 2022-01-27 | 2023-08-03 | Vincent Dert | Determining and/or evaluating a sustainability of a product, a service, an organization and/or a person |
| CN115186507A (en) * | 2022-08-01 | 2022-10-14 | 广东电网有限责任公司电力调度控制中心 | Carbon value prediction method and device based on system dynamics |
| CN116452240A (en) * | 2023-04-13 | 2023-07-18 | 国网北京市电力公司 | Carbon transaction price prediction system, method, equipment and storage medium |
Non-Patent Citations (3)
| Title |
|---|
| 企业碳预算体系构建探究;闫华红等;会计之友;20180108(02);6-10 * |
| 张影等.《预测与评价》.天津大学出版社,2015,(第1版),89-92. * |
| 贾怀勤主编.《数据、模型与决策》.对外经济贸易大学出版社,2012,(第3版),8. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117635202A (en) | 2024-03-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110046755A (en) | Allot the determination method and apparatus in warehouse | |
| CN113743971B (en) | Data processing method and device | |
| US20220083842A1 (en) | Optimal policy learning and recommendation for distribution task using deep reinforcement learning model | |
| CN111008724A (en) | Price prediction method and device, electronic equipment and readable storage medium | |
| CN111889524A (en) | Cold rolling force prediction method and system based on machine learning method | |
| CN116128135A (en) | Data processing method and device, electronic equipment and storage medium | |
| CN114049162A (en) | Model training method, demand prediction method, apparatus, device, and storage medium | |
| CN117635202B (en) | Carbon asset price prediction method, device, electronic equipment and storage medium | |
| CN114022221A (en) | Sales prediction method, acquisition method and device of model thereof, and electronic equipment | |
| CN112231299B (en) | Method and device for dynamically adjusting feature library | |
| CN113705929A (en) | Spring festival holiday load prediction method based on load characteristic curve and typical characteristic value fusion | |
| CN117237085A (en) | Information processing method, equipment and computer readable storage medium | |
| CN116911574A (en) | Three-level supply chain optimization method and device based on whale algorithm and random forest | |
| CN116071102A (en) | Order-based power transformer supply early warning method and system | |
| CN116823127A (en) | Article warehouse-in processing method and device, electronic equipment and storage medium | |
| CN114638405A (en) | Energy demand prediction method and device and computer equipment | |
| CN114549032A (en) | Stock processing method and device, computer storage medium and electronic equipment | |
| CN117350517B (en) | Control method, system, equipment and storage medium for chemical production flow | |
| CN116993028B (en) | Workshop scheduling method and device, storage medium and electronic equipment | |
| CN115712243B (en) | Correction method, device, equipment and medium | |
| CN116862561B (en) | Product heat analysis method and system based on convolutional neural network | |
| CN118035681A (en) | Analysis method and device of carbon emission influence factors, electronic equipment and storage medium | |
| CN117196688A (en) | Price trend prediction method, device, equipment and medium method | |
| CN115759315A (en) | Information processing method, information processing apparatus, electronic device, information processing medium, and program product | |
| CN116823334A (en) | Stock market prediction system and control method thereof |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |