Middle China region economic prediction method and system based on power consumption and economic output data

Abstract

The invention belongs to but is not limited to the technical field of regional economic prediction, and discloses a central China region economic prediction method based on power consumption data and economic output data. Carrying out integration processing on the existing data, and carrying out prediction by adopting an MIDAS (m, k, h)-AR (p) autoregression model; the economic output of the subdivided industries is predicted based on the low-frequency economic output data and the high-frequency power consumption data of the 45 subdivided industries of agriculture and industry; based on the low-frequency economic output data and the high-frequency power consumption data of the nine industries, the economic output of the nine industries is predicted, and the annual regional total output can be obtained through totaling; and predicting the regional economic output based on the regional low-frequency economic output data and the high-frequency power consumption data. According to the method, the model fitting analysis of the low-frequency economic data and the high-frequency power data is constructed, so that the regional economic development is predicted.

Description

Technical Field

[0001] The present invention belongs to but is not limited to the technical field of regional economic forecasting, and in particular relates to a method and system for forecasting the central China regional economy based on power consumption data and economic output data. Background Art

[0002] Electricity is an important foundation for economic development. Through the correlation analysis of electricity data and economic data, we can reveal the trends and characteristics of economic development and provide an important reference for predicting future economic development. Electricity data not only reflects the intensity and scale of economic activities, but also reveals changes in industrial structure and energy utilization. Therefore, it is widely used in economic forecasting and evaluation.

[0003] "Looking at the economy from the perspective of electricity" is not only a data correlation analysis method, but also an important way to gain insight into the laws of economic development. Actively carrying out relevant practices on the issue of "Looking at the economy from the perspective of electricity" will help achieve sustainable economic growth and high-quality development.

[0004] Compared with traditional economic data, electricity consumption data has the unique advantages of real-time dynamics, accuracy, objectivity and leading edge. Through the changes in terminal electricity consumption information, it can fully reflect the economic activity of market entities, timely judge the production and development trends of the industry, and serve local economic decision-making. Therefore, a method for economic forecasting in the central China region based on electricity consumption data and economic output data is needed.

[0005] In the relevant practice of "seeing the economy through electricity", although electricity consumption data has unique advantages such as real-time dynamics, accuracy, objectivity and precedence, the existing technology still faces some technical problems and challenges in the actual application process, mainly including the following aspects:

[0006] 1. Data integration and standardization issues: Power consumption data comes from a wide range of power networks and end users, with various data formats and varying quality. Power data from different regions and time periods may lack unified standards and specifications, making data integration difficult and affecting the accuracy and efficiency of subsequent analysis.

[0007] 2. Data quality and cleaning problems: There may be a lot of noise, outliers and missing values ​​in electricity consumption data. These data quality issues will directly affect the reliability of the analysis results. How to effectively clean and preprocess data to ensure the integrity and accuracy of the data is a major challenge facing existing technologies.

[0008] 3. Model construction and algorithm selection: The relationship between electricity consumption data and economic output data is complex and changeable, and it is necessary to build appropriate mathematical models and algorithms to capture this relationship. However, existing models and algorithms may not be fully adapted to all situations, resulting in biased analysis results. In addition, as the amount of data increases and the complexity increases, the computational efficiency and scalability of the model have also become issues that need to be addressed.

[0009] 4. Real-time and dynamic issues: Electricity consumption data is real-time and dynamic, requiring the analysis system to be able to process and analyze data in real time and promptly reflect the economic activity of market entities. However, existing technologies may be insufficient in real-time data processing and dynamic analysis, making it difficult to meet actual needs.

[0010] 5. Cross-domain knowledge integration: The correlation analysis between electricity consumption data and economic output data involves knowledge from multiple fields, such as economics, statistics, computer science, etc. How to effectively integrate knowledge from different fields to form a comprehensive analysis framework and method is one of the problems that existing technologies need to solve.

[0011] 6. Uncertainty in policies and market environment: The relationship between electricity consumption data and economic output data is affected by many factors, including policy adjustments, market fluctuations, etc. These uncertain factors increase the difficulty and complexity of analysis, requiring the analysis system to have a certain degree of flexibility and adaptability. Summary of the invention

[0012] In view of the problems existing in the prior art, the present invention provides a method and system for economic forecasting in Central China based on power consumption data and economic output data.

[0013] The present invention is implemented as follows: a method for economic forecasting in the Central China region based on power consumption data and economic output data, comprising the following steps:

[0014] Step 1: Data preprocessing: Aggregate the existing data according to the consistency of power data and economic data industry division;

[0015] Step 2: Use the economic output forecasting model of subdivided industries based on time series algorithm to make forecasts;

[0016] Step three, use the regional economic output forecasting model based on time series algorithm to make forecasts.

[0017] Furthermore, the step 2 is based on the low-frequency economic output data (annual output value or operating income data) and high-frequency electricity consumption data (monthly data) of 45 agricultural and industrial sub-sectors, and uses the MIDAS (m, k, h)-AR (p) autoregression model to predict the economic output of the sub-sectors.

[0018] Furthermore, the general form of the adopted MIDAS(m,k,h)-AR(p) autoregressive model is:

[0019]

[0020] Among them, y t is the low-frequency explained variable; is a high-frequency explanatory variable; m is the multiple between high-frequency and low-frequency data: if y t is quarterly data, x t If it is monthly data, then m=3; W(L 1 / m ; θ) is the weighted lag polynomial, L 1 / m is the high frequency lag operator, The form of w(k:θ) is determined by the specific weight function chosen; (1-λL p ) is a dynamic autoregressive structure added to high-frequency data to offset the "pseudo-seasonal effect" produced by the autoregressive lag term of low-frequency data; h is the number of advance steps. If the high-frequency data is monthly data, it represents the month in advance. When h is less than the frequency m of the high-frequency data, the model can perform instant prediction.

[0021] Furthermore, based on the low-frequency economic output data (annual value-added data) and high-frequency electricity consumption data (monthly data) of the nine industries, the MIDAS (m, k, h)-AR (p) autoregression model is used for prediction to obtain the predicted values ​​of the added value of the nine industries, and the annual regional total output can be obtained by adding them up.

[0022] Furthermore, the step three also uses the MIDAS (m, k, h)-AR (p) autoregressive model to predict regional economic output based on regional low-frequency economic output data (quarterly GDP data) and high-frequency electricity consumption data (monthly data).

[0023] Another object of the present invention is to provide a method for predicting the economic situation of the Central China region using power consumption data and economic output data, a system for predicting the economic situation of the Central China region using power consumption data and economic output data, comprising:

[0024] Data preprocessing module, used to aggregate existing data according to the consistency of power data and economic data industry division;

[0025] The sub-industry economic output forecasting module is used to make forecasts using the sub-industry economic output forecasting model based on the time series algorithm;

[0026] The regional economic output forecasting module is used to make forecasts using the regional economic output forecasting model based on the time series algorithm.

[0027] Another object of the present invention is to provide a computer device, which includes a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the method for economic forecasting in the central China region based on electricity consumption data and economic output data.

[0028] Another object of the present invention is to provide a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the processor executes the steps of the method for economic forecasting in the central China region based on electricity consumption data and economic output data.

[0029] Another object of the present invention is to provide an information data processing terminal, which includes the central China regional economic forecasting system containing the power consumption data and economic output data.

[0030] In combination with the above technical solutions and the technical problems solved, the advantages and positive effects of the technical solutions to be protected by the present invention are as follows:

[0031] First, the present invention collects power consumption data and economic output data from power grid companies, studies the calculation of regional economic development based on power data, and predicts regional economic development by constructing a model fitting analysis of low-frequency economic data and high-frequency power data. The present invention constructs an economic output prediction model based on real-time and advanced power consumption data to achieve the effect of "seeing the economy through power".

[0032] Second, the present invention proposes significant improvements and innovations to the technical problems of existing power consumption data and economic output data in regional economic forecasting, which are specifically reflected in the following aspects:

[0033] First, the present invention achieves consistency in industry classification between power data and economic data through data preprocessing, effectively solving the problem of data integration and standardization. This step improves the comparability and analysis efficiency of data, and provides a solid foundation for subsequent economic forecasting.

[0034] Secondly, the present invention adopts the MIDAS (m, k, h)-AR (p) autoregressive model based on time series algorithm when predicting the economic output of subdivided industries. This model can make full use of the information of low-frequency economic output data and high-frequency electricity consumption data, and improve the accuracy and reliability of prediction. Compared with the prior art, the present invention has made significant progress in model construction and algorithm selection.

[0035] Furthermore, the present invention not only focuses on the economic output forecast of subdivided industries, but also further forecasts the regional economic output. By applying the same MIDAS (m, k, h)-AR (p) autoregressive model, the present invention achieves accurate forecasts of the economic output of nine industries and the entire region. This step enhances the practicality and application value of the present invention.

[0036] Finally, the present invention solves the shortcomings of the existing technology in real-time and dynamic analysis by introducing high-frequency electricity consumption data as an important variable for prediction. The real-time dynamic characteristics of high-frequency electricity consumption data enable the present invention to timely reflect the level of economic activity of market entities, provide governments and enterprises with more timely and accurate economic forecasting information, and contribute to the realization of sustainable economic growth and high-quality development. In summary, the present invention has achieved significant technical progress and application value in the regional economic forecasting of electricity consumption data and economic output data. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 It is a flow chart of a method for economic forecasting in the Central China region based on power consumption data and economic output data provided by an embodiment of the present invention;

[0038] Figure 2 It is a structural diagram of the Central China regional economic forecasting system for power consumption data and economic output data provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0039] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0040] like Figure 1 As shown, the method for economic forecasting in Central China based on power consumption data and economic output data provided by the embodiment of the present invention includes:

[0041] Step 1: Data preprocessing: Aggregate the existing data according to the consistency of power data and economic data industry division;

[0042] Step 2: Use the economic output forecasting model of subdivided industries based on time series algorithm to make forecasts;

[0043] Step three, use the regional economic output forecasting model based on time series algorithm to make forecasts.

[0044] The embodiment of the present invention performs preliminary data processing and analysis on data from the power grid, the statistical bureau, and the national research network, and combines the determination of relevant models such as subdivided industry economic forecasts and regional economic forecasts in subsequent research to perform economic output forecasts.

[0045] First, data preprocessing

[0046] The present invention sorts and summarizes the electricity consumption and economic data of each sub-industry in the four provinces in central China, and obtains the monthly electricity consumption data from 2018 to 2023, the annual economic data of the sub-industry from 2018 to 2022, and the regional economic quarterly data from 2018Q1 to 2023Q3. In order to ensure the industry unification of the four provinces, the present invention finally selects 45 sub-industries and 9 major industries for analysis in combination with the economic data disclosed by each province. Among them, in the selection of economic data, combined with the data disclosed in the statistical yearbooks of the four provinces in central China, the present invention selects the gross output value data of the sub-industries under agriculture, forestry, animal husbandry and fishery for analysis; selects the operating income data of the sub-industries under industry for analysis; selects the added value data for agriculture, forestry, animal husbandry, fishery, industry, construction, transportation, warehousing and postal industry, wholesale and retail industry, accommodation and catering industry, finance, real estate industry, and other service industries (including information transmission, software and information technology services, leasing and business services, public services and management organizations) for analysis. Regional economic data selects GDP data for analysis.

[0047] Table 1

[0048] Serial number data Data granularity Data Source 1 Power consumption Province, Month Central China Branch 2 Economic output by industry Province, year Provincial Statistics Bureau 3 Value added by industry Province, year National Research Network 4 GDP Province, Season EPS Database

[0049] Second, the economic output forecasting model of sub-industries based on time series algorithm

[0050] When selecting a time series prediction method, the present invention takes into account the needs of different data frequencies, and finally adopts a mixed frequency processing method (MIDAS). This method does not need to convert different frequency data before modeling, and can directly use different frequency variable data to establish an econometric model. The emergence of mixed frequency data models fills the shortcomings of traditional same-frequency econometric models. Its analysis framework is more flexible, and high-frequency and low-frequency time series data can be directly introduced into the same model, thereby capturing more effective information of high-frequency data, and maximizing the use of this information, avoiding the loss of sample information by converting variables to the same frequency. The basic principle of MIDAS is that low-frequency variables are used as explained variables, and high-frequency data corresponding to the same period are used as explanatory variables.

[0051] The present invention is based on low-frequency economic output data (annual output value or operating income data) and high-frequency electricity consumption data (monthly data) of 45 sub-sectors in agriculture and industry, and uses the MIDAS (m, k, h)-AR (p) autoregression model to predict the economic output of the sub-sectors. In the MIDAS (m, k, h)-AR (p) autoregression model: AP is autoregression, p is the autoregression term of the low-frequency data; m is the ratio of high-frequency data to low-frequency data, k is the optimal lag order of high-frequency data, and h is the number of advance steps. The principle is that the low-frequency variable is used as the explained variable, and the high-frequency data corresponding to the same period is used as the explanatory variable. The form of the regression equation is:

[0052]

[0053] Data stationarity test. When modeling mixed frequency data, in order to ensure the accuracy and precision of the model, it is necessary to test the stationarity of the time series. The ADF test (Augmented Dickey-Fuller Test) is used to test the stationarity of the time series. If the p value is less than 0.05, it means that the null hypothesis is rejected and the time series is a stationary time series. Otherwise, it means that the time series is a non-stationary series. For non-stationary time series, differential processing is performed to make it stationary.

[0054] The weight function and lag order are determined. In Eviews, the weight functions include Step, PDL / Almon, Exponential Almon, U-MIDAS, Beta, etc. Among them, Step and PDL / Almon can automatically select the lag order, and other weight functions can customize the order. By constructing the MIDAS (m, k) model, in order to avoid the problem of missing degrees of freedom caused by introducing more lag terms, the longest lag order of high-frequency variables is set to 12. In order to obtain a better fitting effect, the weight function of PDL / Almon was selected after multiple simulation attempts. In the case of the optimal weight polynomial, the MIDAS (m, k) model is established according to each variable to determine the optimal lag order of the model. Based on this, forward prediction and autoregressive terms are added respectively to establish the MIDAS (m, k, h) -AR (p) autoregressive model for real-time forecasting and short-term forecasting of economic output.

[0055] Estimation and prediction. After determining the parameters m, k, h, and p, the accuracy on the training data set can be estimated using the MIDAS(m, k, h)-AR(p) autoregressive model, and the fitted model can be used to predict the values ​​of the test data to verify the deviation from the actual values ​​and the accuracy of the model.

[0056] Third, regional economic output forecasting model based on time series algorithm

[0057] The present invention adopts two methods to calculate regional economic output.

[0058] (1) Based on the low-frequency economic output data (annual value-added data) and high-frequency electricity consumption data (monthly data) of the nine industries, the MIDAS (m, k, h)-AR (p) autoregressive model is used for forecasting. According to the above steps, the economic output of the nine industries is forecasted to obtain the forecast value of the added value of the nine industries, and the total annual regional output can be obtained by summing them up.

[0059] (2) Based on the regional low-frequency economic output data (quarterly GDP data) and high-frequency electricity consumption data (monthly data), the MIDAS (m, k, h)-AR (p) autoregressive model is used for forecasting. The regional economic output is forecasted according to the above steps.

[0060] like Figure 2 As shown, an embodiment of the present invention provides a method for predicting the economic situation in Central China based on power consumption data and economic output data, a system for predicting the economic situation in Central China based on power consumption data and economic output data, including:

[0061] Data preprocessing module, used to aggregate existing data according to the consistency of power data and economic data industry division;

[0062] The sub-industry economic output forecasting module is used to make forecasts using the sub-industry economic output forecasting model based on the time series algorithm;

[0063] The regional economic output forecasting module is used to make forecasts using the regional economic output forecasting model based on the time series algorithm.

[0064] Example: Total output forecast for nine industries and regions

[0065] Step 1: Data preprocessing

[0066] 1. Collect electricity consumption data and economic output data of nine major industries in Central China.

[0067] 2. Aggregate electricity consumption data and economic output data by nine major industries to ensure consistency in industry classification.

[0068] Step 3: Regional Economic Output Forecast

[0069] 1. Based on the low-frequency economic output data and high-frequency electricity consumption data of nine industries, the MIDAS (m, k, h)-AR (p) autoregressive model is used for prediction.

[0070] 2. Data refinement:

[0071] -Low-frequency economic output data: annual value-added data.

[0072] -High-frequency electricity consumption data: monthly data.

[0073] 3. Use historical data to train models and predict the added value of nine industries.

[0074] 4. Add up the forecasts of value added of the nine industries to arrive at the total annual regional output.

[0075] Step 3 Extension: Regional Quarterly GDP Forecast

[0076] 1. Based on the regional low-frequency economic output data and high-frequency electricity consumption data, the MIDAS(m,k,h)-AR(p) autoregressive model is used to predict regional economic output.

[0077] 2. Data refinement:

[0078] -Low-frequency economic output data: quarterly GDP data.

[0079] -High-frequency electricity consumption data: monthly data.

[0080] 3. The model formula is the same as that in Example 1.

[0081] 4. Use historical data to train the model and predict regional GDP for future quarters.

[0082] Through this embodiment, by utilizing electricity consumption data and economic output data, accurate forecasts of subdivided industries and overall economic output in the central China region can be achieved, thus improving the scientific nature and accuracy of economic decision-making.

[0083] It should be noted that the embodiments of the present invention can be implemented by hardware, software, or a combination of software and hardware. The hardware part can be implemented using dedicated logic; the software part can be stored in a memory and executed by an appropriate instruction execution system, such as a microprocessor or dedicated design hardware. It can be understood by a person of ordinary skill in the art that the above-mentioned devices and methods can be implemented using computer executable instructions and / or contained in a processor control code, such as a carrier medium such as a disk, CD or DVD-ROM, a programmable memory such as a read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. Such code is provided on the carrier medium. The device and its modules of the present invention can be implemented by hardware circuits such as very large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, etc., or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., can also be implemented by software executed by various types of processors, and can also be implemented by a combination of the above-mentioned hardware circuits and software, such as firmware.

[0084] The above description is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any modifications, equivalent substitutions and improvements made by any technician familiar with the technical field within the technical scope disclosed by the present invention and within the spirit and principle of the present invention should be covered by the protection scope of the present invention.

Claims

1. A method for economic forecasting in Central China based on power consumption data and economic output data. It is characterized in that The following steps are involved: Step 1: Data preprocessing: Aggregate the existing data according to the consistency of power data and economic data industry division; Step 2: Use the economic output forecasting model of subdivided industries based on time series algorithm to forecast the economic output of subdivided industries; Step three, use the regional economic output forecasting model based on time series algorithm to predict regional economic output.

2. The method for economic forecasting in Central China based on power consumption data and economic output data as claimed in claim 1, It is characterized in that The second step is based on the low-frequency economic output data and high-frequency electricity consumption data of 45 agricultural and industrial sub-sectors, and uses the MIDAS (m, k, h)-AR (p) autoregression model to predict the economic output of the sub-sectors; the low-frequency economic output data is annual output value or operating income data, and the high-frequency electricity consumption data is monthly data.

3. The method for economic forecasting in Central China based on power consumption data and economic output data as claimed in claim 2, It is characterized in that The formula of the MIDAS(m,k,h)-AR(p) autoregressive model used is: Among them, y t is the low-frequency explained variable; is a high-frequency explanatory variable; m is the multiple between high-frequency and low-frequency data: if y t is quarterly data, x t If it is monthly data, then m=3; W(L 1 / m ; θ) is the weighted lag polynomial, L 1 / m is the high frequency lag operator, The form of w(k:θ) is determined by the specific weight function chosen; (1-λL p ) is a dynamic autoregressive structure added to high-frequency data to offset the "pseudo-seasonal effect" produced by the autoregressive lag term of low-frequency data; h is the number of advance steps. If the high-frequency data is monthly data, it represents the month in advance. When h is less than the frequency m of the high-frequency data, the model can perform instant prediction.

4. The method for economic forecasting in Central China based on power consumption data and economic output data as claimed in claim 1, It is characterized in that The step three uses the MIDAS (m, k, h)-AR (p) autoregressive model to make predictions based on the low-frequency economic output data and high-frequency electricity consumption data of the nine industries, and obtains the predicted values ​​of the added value of the nine industries, which can be summed up to obtain the annual regional total output; the low-frequency economic output data is the annual added value data, and the high-frequency electricity consumption data is the monthly data.

5. The method for economic forecasting in Central China based on power consumption data and economic output data as claimed in claim 1, It is characterized in that The step three also uses the MIDAS (m, k, h)-AR (p) autoregressive model to predict regional economic output based on regional low-frequency economic output data and high-frequency electricity consumption data; the regional low-frequency economic output data is quarterly GDP data, and the high-frequency electricity consumption data is monthly data.

6. A system for forecasting the economic situation in Central China using power consumption data and economic output data according to the method for forecasting the economic situation in Central China using power consumption data and economic output data as claimed in any one of claims 1 to 5, It is characterized in that include: Data preprocessing module, used to aggregate existing data according to the consistency of power data and economic data industry division; The sub-industry economic output forecasting module is used to make forecasts using the sub-industry economic output forecasting model based on the time series algorithm; The regional economic output forecasting module is used to make forecasts using the regional economic output forecasting model based on the time series algorithm.

7. A computer device, It is characterized in that The computer device includes a memory and a processor, the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the method for economic forecasting in the central China region based on power consumption data and economic output data as described in any one of claims 1 to 5.

8. A computer-readable storage medium storing a computer program, which, when executed by a processor, enables the processor to execute the steps of the method for economic forecasting in Central China based on power consumption data and economic output data as described in any one of claims 1 to 5.

9. An information data processing terminal, It is characterized in that The information data processing terminal includes the Central China regional economic forecasting system of the power consumption data and economic output data as described in claim 6.

Images