CN110110885A - Information forecasting method, device, computer equipment and storage medium - Google Patents

Abstract

The present application relates to the field of big data processing, in particular to an information prediction method, device, computer equipment and storage medium. The method includes: obtaining the quarterly growth rate value of GDP at current prices, and obtaining the basic index data corresponding to each basic index one by one from the basic index database; extracting multiple time-frequency basic index sequences from each basic index data to obtain multiple groups of basic indicators Sequence; calculate the correlation coefficient between each group of basic index sequence and the growth rate value respectively; select the basic index sequence associated with the growth rate value in each group according to the correlation coefficient as the downstream correlation index sequence; perform indexing on each group of downstream correlation index sequence The components are analyzed and constructed to obtain multiple real economy index forecast models; the real economy index is forecasted for the real economy based on multiple real economy index forecast models. The present invention predicts the real economy based on various basic indicators in the real economy scene, which is scientific, reliable and more accurate.

Description

Information prediction method, device, computer equipment and storage medium

technical field

The present application relates to the field of computer technology, in particular to an information prediction method, device, computer equipment and storage medium.

Background technique

The real economy directly creates material wealth, which is the concentrated expression of social productivity and the material basis of social wealth and comprehensive national strength. A developed and stable real economy is of great significance to providing jobs, improving people's lives, and achieving sustainable economic development and social stability. Compared with the virtual economy, the real economy often has disadvantages such as higher input costs, longer output cycles, and limited profit margins. Therefore, the real economy needs more attention and support.

Without the support of the real economy, there will be no solid foundation for the return on financial asset investment and transactions. Breaking away from the real economy and over-hyping assets will not only affect economic development and widen the social gap between rich and poor, but will also increase economic, financial and social risks. Therefore, it is necessary to take precautions against the trend of the real economy in advance. However, there is currently a lack of effective means to accurately predict the trend of the real economy.

Contents of the invention

Based on this, it is necessary to provide an information forecasting method, device, computer equipment and storage medium capable of forecasting the real economy index for the real economy in view of the above technical problems.

An information prediction method, the method comprising:

Obtain the quarterly growth rate value of GDP at the current price, and obtain the basic index data corresponding to each basic index one by one from the basic index database;

Extracting a plurality of time-frequency basic index sequences from each of the basic index data to obtain multiple sets of basic index sequences;

Calculate the correlation coefficient between each group of said basic index sequence and said growth rate value respectively;

According to the correlation coefficient, each group of basic index sequences associated with the growth rate value is selected as a downward correlation index sequence;

Carry out index component analysis on the downward correlation index sequence described in each group and construct multiple real economy index prediction models;

The real economy index is predicted for the real economy according to a plurality of real economy index prediction models.

In one of the embodiments, the respectively calculating the correlation coefficient between each set of the basic index sequence and the growth rate value includes:

Bringing the basic index sequence and the growth rate value into the correlation calculation formula for calculation;

The absolute value of the result calculated according to the correlation calculation formula is set as the correlation coefficient.

In one of the embodiments, according to the correlation coefficient, each group of basic index sequences associated with the growth rate value is selected as a downward correlation index sequence, including:

Obtaining a preset correlation threshold, and extracting all basic index sequences with correlation coefficients not less than the preset correlation threshold as index sequences to be processed;

Acquiring attribute information corresponding to the index sequence to be processed, and classifying the index sequence to be processed according to the attribute information;

The basic index sequence with the largest correlation coefficient among the index sequences to be processed in each classification is set as the downward correlation index sequence.

In one of the embodiments, the setting of the basic index sequence with the largest correlation coefficient among the index sequences to be processed in each classification as the downward correlation index sequence includes:

Obtaining the number of classifications of the index sequence to be processed and the lower limit threshold of the number of preset indexes;

Comparing the number of categories with the lower limit threshold of the number of preset indicators;

When the classification quantity is less than the preset index quantity lower limit threshold, calculate the first difference between the preset index quantity lower limit threshold and the classification quantity;

Sorting the correlation coefficients of the remaining index sequences to be processed from large to small, extracting the index sequences to be processed at the top as downstream correlation index sequences, and extracting the index sequences to be processed at the top The number of index sequences is consistent with the first difference.

In one of the embodiments, the index component analysis of the downward correlation index sequence is carried out to obtain the real economy index prediction model, including:

Perform standardization processing on all the downlink related index sequences to obtain a data matrix;

Obtaining the covariance matrix of the downlink correlation index sequence according to the data matrix, and calculating the eigenroot, eigenvector and principal component variance contribution rate of the covariance matrix;

Obtain the preset minimum variance contribution rate threshold;

Screening out principal component expressions whose variance contribution rate of the principal component is greater than the preset minimum variance contribution rate threshold;

A real economy index prediction model is constructed according to the screened principal component expressions.

In one of the embodiments, the construction of the real economy index prediction model based on the screened principal component expressions includes:

Obtain the preset minimum principal component contribution rate threshold;

Comparing the sum of variance contribution rates of all screened principal components with the preset minimum principal component contribution rate threshold;

When the sum of the principal component variance contribution rates is less than the preset minimum principal component contribution rate threshold, calculate a second difference between the two;

Sorting the principal component variance contribution rates of the remaining principal component expressions from large to small, extracting the top principal component expressions, and the sum of the extracted principal component variance contribution rates is not less than the second difference.

An information prediction device, the device comprising:

The value acquisition module is used to obtain the quarterly growth rate value of GDP at the current price, and obtain the basic index data corresponding to each basic index one by one from the basic index database;

A sequence extraction module, configured to extract a plurality of time-frequency basic index sequences from each of the basic index data to obtain multiple sets of basic index sequences;

Calculation module, used to calculate the correlation coefficient between each set of said basic index sequence and said growth rate value respectively;

A screening module, according to the correlation coefficient, screens out each group of basic index sequences associated with the growth rate value as a downward correlation index sequence;

A model construction module, which is used to analyze the index components of each group of the downward correlation index sequences and construct a plurality of real economy index prediction models;

The index prediction module is used for predicting the real economy index of the real economy according to a plurality of said real economy index prediction models.

In one of the embodiments, the screening module includes:

An extraction unit, configured to obtain a preset correlation threshold, and extract a basic index sequence with a correlation coefficient greater than the preset correlation threshold as an index sequence to be processed;

The classification unit is used to obtain the attribute information corresponding to the index sequence to be processed, and classify the index sequence to be processed according to the attribute information;

The setting unit is used to set the basic index sequence with the largest correlation coefficient in each type of index sequence to be processed as the downward correlation index sequence.

A computer device, comprising a memory and a processor, the memory stores a computer program, and the processor implements the following steps when executing the computer program:

Obtain the quarterly growth rate value of GDP at the current price, and obtain the basic index data corresponding to each basic index one by one from the basic index database;

Extracting a plurality of time-frequency basic index sequences from each of the basic index data to obtain multiple sets of basic index sequences;

Calculate the correlation coefficient between each group of said basic index sequence and said growth rate value respectively;

According to the correlation coefficient, each group of basic index sequences associated with the growth rate value is selected as a downward correlation index sequence;

Carry out index component analysis on the downward correlation index sequence described in each group and construct multiple real economy index prediction models;

The real economy index is predicted for the real economy according to a plurality of real economy index prediction models.

A computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

Obtain the quarterly growth rate value of GDP at the current price, and obtain the basic index data corresponding to each basic index one by one from the basic index database;

Extracting a plurality of time-frequency basic index sequences from each of the basic index data to obtain multiple sets of basic index sequences;

Calculate the correlation coefficient between each group of said basic index sequence and said growth rate value respectively;

According to the correlation coefficient, each group of basic index sequences associated with the growth rate value is selected as a downward correlation index sequence;

Carry out index component analysis on the downward correlation index sequence described in each group and construct multiple real economy index prediction models;

The real economy index is predicted for the real economy according to a plurality of real economy index prediction models.

The above-mentioned information prediction method, device, computer equipment and storage medium, by obtaining the index data corresponding to all the basic indexes from the basic index database, and extracting a plurality of time-frequency basic index sequences from each of the basic index data, obtain multiple sets of The basic index sequence, and then determine the downward correlation index according to the correlation coefficient between each basic index sequence and the growth rate value, so as to use the downward correlation index to construct the real economy index forecasting model. The present invention has many indexes and abundant sources, and the constructed downward reserve index prediction model is scientific and reliable. Moreover, the present invention predicts the real economy based on various basic indicators under the real economy, and the accuracy is higher. The present invention also extracts multiple sets of basic index sequences, and predicts the real economy index of the real economy through multiple real economy index prediction models, further reducing the impact caused by changes in policies, situations, and time.

Description of drawings

Fig. 1 is an application scenario diagram of an information prediction method in an embodiment;

Fig. 2 is a schematic flow chart of an information prediction method in an embodiment;

FIG. 3 is a schematic flow diagram of setting a downlink correlation index sequence in an embodiment;

Fig. 4 is the schematic flow chart of constructing real economy index prediction model in another embodiment;

Fig. 5 is the schematic flow chart of constructing real economy index prediction model in another embodiment;

Fig. 6 is the schematic diagram of the 4 period real economy index prediction models that construct in an embodiment;

Fig. 7 is a structural block diagram of an information prediction device in an embodiment;

Figure 8 is a diagram of the internal structure of a computer device in one embodiment.

Detailed ways

In order to make the purpose, technical solution and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain the present application, and are not intended to limit the present application.

The information prediction method provided in this application can be applied to the application environment shown in FIG. 1 . Wherein, the terminal 102 communicates with the server 104 through the network. The user can send a request for the real economic index of the expected forecast period to the server 104 through the terminal 102; after receiving the request, the server 104 predicts the real economic index of the forecast period, and sends the predicted real economic index to the terminal 102 through the network; The terminal 102 displays the predicted real economic index and lets the user know it. Wherein, the terminal 102 can be, but not limited to, various personal computers, notebook computers, smart phones, tablet computers and portable wearable devices, and the server 104 can be realized by an independent server or a server cluster composed of multiple servers.

In one embodiment, as shown in FIG. 2 , an information prediction method is provided. The method is applied to the server in FIG. 1 as an example for illustration, including the following steps:

Step 202, obtain the quarterly growth rate value of GDP at the current price, and obtain the basic index data corresponding to each basic index one by one from the basic index database.

In this embodiment, the basic indicators in the basic indicator library are derived from more than 2,000 macroeconomic indicators, basically covering important indicators of the entire macro economy. These indicators include non-agricultural employment population, employment rate, unemployment rate, balance of payments, gross national product, producer price index, consumer price index, disposable personal income, personal consumption expenditure, balance of savings deposits of urban and rural residents, investment indicators, financial Indicators (financial indicators include interest rate, exchange rate, money supply, deposit and loan balance of financial institutions, total financial assets, etc.), consumer confidence index, purchasing managers index, durable goods orders, industrial production index, equipment utilization rate, retail sales index , Total Retail Sales of Consumer Goods, Consumer Credit, Housing Starts and Building Permits, Construction Expenditures, Producer Price Index, Wholesale Price Index, Foreign Trade, Factory Orders, Durable Goods Orders, Current Account, and Commercial Inventories, etc.

Each basic indicator in the basic indicator database has corresponding basic indicator data, and it is updated and stored regularly. All basic indicators are updated and stored at least once a year, but some basic indicators will not only store annual data once a year, but also obtain multiple update data according to the situation, such as exchange rates, money supply, etc. updated on a daily basis, and weekly The updated storage includes production price index, industrial production index, retail sales index, etc., the monthly updated storage includes non-agricultural employment population, employment rate, unemployment rate, financial institution deposit and loan balance, etc., and the quarterly updated storage includes investment Indicators, Durable Goods Orders, Equipment Utilization, Factory Orders, and more.

The server 104 acquires the quarterly growth rate value of GDP at the current price within a preset period, and acquires the basic indicator data of multiple basic indicators within the same preset period one by one. Taking the basic indicator database as an example, which stores 2000 basic indicators, the data of these basic indicators constitute a data sequence of 2000 basic indicators. The greater the number of growth rates in the preset period and the longer the data series, the higher the accuracy of the constructed real economy index forecast model. Therefore, in order to improve the accuracy of the real economy index forecast model, each basic index in the preset period corresponds to The number of data is generally not less than 5.

Step 204, extract multiple time-frequency basic index sequences from each basic index data to obtain multiple sets of basic index sequences.

Different basic index sequences are extracted from the basic index data according to different time frequencies within a predetermined period to obtain multiple sets of basic index sequences. The time frequency of extraction can be daily, weekly, monthly, quarterly, etc. For example, when extracting the basic index series of the exchange rate, the server will extract the exchange rate index series updated daily, the exchange rate index series updated weekly, the exchange rate index series updated monthly, and the exchange rate index series updated quarterly from the basic index library. The series of exchange rate indicators and the series of exchange rate indicators updated annually.

In order to facilitate the construction of the subsequent real economy index forecasting model, the basic index sequence of all basic index data is uniformly extracted within a predetermined period according to the time frequency. If the minimum update frequency corresponding to the basic indicator is greater than the time frequency, set the indicator data of the previous time period as the indicator data of this time period. For example, GNP is the final result of the primary income distribution of all resident units in a country (or region) within a certain period (usually one year), so there is no daily update data for GNP. When extracting the daily updated GNP index series in 2017, set the GNP in 2016 as the daily updated GNP in 2017.

Step 206, respectively calculate the correlation coefficient between each group of basic index sequence and the growth rate value.

The server calculates the correlation coefficient between the basic index series and the growth rate value respectively. When calculating the correlation coefficient, the server uses correlation calculation formulas such as Pearson correlation coefficient or Spearman correlation coefficient to perform the calculation. The server can substitute the basic index sequence and growth rate value into the correlation calculation formula for calculation, and directly set the calculation result as the correlation coefficient.

For example, the server can use the calculation formula of Pearson correlation coefficient:

In the above formula, X is the basic index sequence, Y is the quarterly growth rate of current GDP, and ρ _{(X, Y)} is the correlation coefficient. The Pearson correlation coefficient ρ _{(X,Y) of two continuous variables (X,Y)} is equal to the covariance cov(X,Y) between them divided by the product of their respective standard deviations (σ _X ,σ _Y ). The value of the coefficient is always between -1.0 and 1.0. Variables close to 0 are said to have no correlation, and variables close to 1 or -1 are said to have strong correlation.

The server can also use the Spearman correlation coefficient calculation formula:

When the server calculates the correlation coefficient, X is the basic index sequence, Y is the quarterly growth rate of current GDP, and ρ is the correlation coefficient calculated by the server. The dependent variable in the model is the growth rate of the real economy, and the independent variables are multiple downward correlation index sequences, each of which is given different weights.

In one embodiment, calculating the correlation coefficient between each basic index sequence and the growth rate value includes: substituting the basic index sequence and the growth rate value into the correlation calculation formula for calculation; calculating the result obtained according to the correlation calculation formula The absolute value of is set as the correlation coefficient.

Specifically, the basic index sequence and growth rate values can be brought into the Pearson correlation coefficient calculation formula or the Spearman correlation coefficient calculation formula for calculation, but the absolute value of all calculation results is set as the correlation coefficient.

In step 208, according to the correlation coefficient, each group of basic index sequences associated with the growth rate value is selected as the downward correlation index sequence.

According to the preset index screening rules, the server selects each group of basic index sequences that meet the forecasting needs of the real economy from 2000 basic index sequences according to the correlation coefficient at the same time and frequency as the downward correlation index sequence. The preset index screening rules can extract all the basic index sequences whose correlation coefficients fall within the screening range as the downstream correlation index sequences, or other selection rules. The series of downward correlation indicators screened out at different time frequencies may change.

Step 210, analyzing the index components of each group of downward correlation index sequences and constructing multiple real economy index forecasting models.

Perform principal component analysis on the selected groups of downward correlation indicators. Each group of downstream correlation indicators obtains multiple principal component expressions according to the results of principal component analysis, and then filters the principal component expressions. The expression constructs the real economy index forecasting model. In the process, the real economy index forecast models constructed according to different time frequencies are not exactly the same.

Step 212: Predict the real economy index for the real economy according to multiple real economy index forecasting models.

By substituting the data series of each downward correlation index at the current moment into multiple real economy index forecasting models, multiple real economy index data at the next moment can be predicted. Since the real economy index is updated according to policy, situation and other events and the historical data of each basic index, when the real economy index forecast model is constructed, the server can substitute the current data sequence into the real economy index forecast model to get the next Forecast data of the real economic index at the moment. For example, the server can obtain the forecast data of the real economy index in the second period (second quarter) of 2018 by substituting the data sequence of the first period (first quarter) of 2018 into the real economic index forecasting model.

The server can analyze multiple forecasted real economic indexes, such as generating real economic trend charts and corresponding warning prompts, or sending forecast results to the terminal for users to make decisions.

In the above information forecasting method, the index data corresponding to all basic indicators are obtained from the basic index database, and the downward correlation index sequence is determined according to the correlation coefficient between each basic index sequence and the growth rate value, and then the downward correlation index sequence is used to construct the real economy. Exponential forecasting model. The above-mentioned indicators are numerous and rich in sources, so the real economy index forecasting model constructed is scientific and reliable. Moreover, the above-mentioned information prediction method is based on various basic indicators in the real economy scenario to predict the real economy. Not only is the accuracy higher, but also because the number of data in each data series is generally not less than 5, it can reduce the risk caused by policies and situations. and the effects of changes over time. Moreover, this method also extracts multiple groups of basic index sequences, and predicts the real economy index of the real economy through multiple real economy index prediction models, further reducing the impact caused by changes in policies, situations, and time.

In one embodiment, as shown in FIG. 3 , according to the correlation coefficient, the basic index sequence associated with the growth rate value of the quarterly GDP at the current price is selected as the downward correlation index sequence for each group, which specifically includes the following steps:

Step 302, acquiring a preset correlation threshold, and extracting basic index sequences with correlation coefficients greater than the preset correlation threshold as index sequences to be processed.

The server obtains the preset correlation thresholds from the database, and compares the correlation coefficients calculated according to each basic index sequence with the preset correlation thresholds. When the server determines that the correlation coefficients are greater than the preset correlation thresholds, the corresponding basic Index sequences are extracted as pending index sequences.

Step 304, acquiring attribute information corresponding to the index sequence to be processed, and classifying the index sequence to be processed according to the attribute information.

Because basic indicators cover all aspects of the macro economy, there are situations where multiple basic macro indicators can reflect a class of economic issues. Classify and mark each basic index in advance, and assign different attribute information. The server respectively determines the indicators to be processed according to the sequence of indicators to be processed, and then obtains the corresponding attribute information from the basic indicator database according to the indicators to be processed, and divides the basic indicators belonging to the same attribute information into the same category.

For example, attribute information corresponding to non-agricultural employment population, employment rate, and unemployment rate are all classified as non-agricultural work data; attribute information corresponding to gross national product, gross domestic product, industrial production index, and factory orders are all classified as production data ; Corresponding attribute information such as producer price index, consumer price index, retail sales index, and total retail sales of social consumer goods are all divided into sales data.

Step 306, setting the basic index sequence with the largest correlation coefficient in each type of index sequence to be processed as the downlink correlation index sequence.

Specifically, when the time period is T1, the server classifies all the index sequences to be processed according to the attribute information, and divides them into M categories in total. Type M data are non-agricultural work data, production data, and sales data. Among them, the correlation coefficients corresponding to the non-agricultural employment population, employment rate, and unemployment rate in the non-agricultural work data are 0.6, 0.8, and 0.7 respectively, and the correlation coefficient corresponding to the employment rate index is the largest. Therefore, the server uses the correlation coefficient of the non-agricultural work data The maximum employment rate is set as a downward correlation indicator series.

In the above information prediction method, each basic index is classified and marked in advance, and the basic index sequence with the largest correlation coefficient in each type of index sequence to be processed is set as the downward correlation index sequence to ensure that each type of index is selected. The diversification of the index sequence to be processed is realized, which not only reduces the impact caused by changes in policies, situations, and time, but also further improves the accuracy of the model.

In one embodiment, after setting the basic index sequence with the largest correlation coefficient in each type of index sequence to be processed as the downward correlation index sequence, the method further includes:

In step 308, the number of classifications of the index sequence to be processed and the lower limit threshold of the number of preset indexes are acquired.

The server obtains the preset lower limit threshold of the number of indicators from the database, and directly obtains the classification quantity of the indicator sequence to be processed. The lower limit threshold of the number of preset indicators refers to the minimum number value set to ensure the accuracy of the final forecast model of the real economy index. Generally, the lower limit threshold of the number of preset indicators is not less than 10.

Step 310, comparing the number of categories with the lower limit threshold of the number of preset indicators.

The server compares the number of classifications with the lower limit threshold of the number of preset indicators. When the number of classifications is not less than the lower limit threshold of the number of preset indicators, it can reduce the impact caused by the dimensionality reduction of the principal component analysis method and ensure the final forecast of the real economy index. model accuracy.

Step 312, when the number of classifications is less than the preset lower limit threshold of the number of indicators, calculate a first difference between the lower limit threshold of the number of preset indicators and the number of classifications.

When the number of classification categories is less than the preset lower limit threshold of the number of indicators, calculate the first difference Q between the preset lower limit threshold of the number of indicators and the number of classification categories.

Step 314, sort the correlation coefficients of the remaining index sequences to be processed from large to small, and extract the top index to be processed as a downlink correlation index, and the number of the extracted top index to be processed is the same as the number of A difference is the same.

The correlation coefficients of the remaining extracted basic index sequences are sorted from large to small, and the top Q-ranked basic index sequences are also extracted as downward correlation index sequences.

In the above-mentioned information prediction method, the lower limit threshold of the number of preset indicators is determined in advance, which reduces the impact caused by the dimensionality reduction of the principal component analysis method and ensures the accuracy of the final real economy index prediction model constructed. When the number of classifications is less than the lower limit threshold of the preset number of indicators, extracting the pending indicators with the highest correlation coefficient values not only makes up for the impact caused by the lack of quantity, but also each pending indicator is closely related to the real economic data, ensuring This ensures the accuracy of the real economy index forecasting model.

In one embodiment, as shown in Figure 4, the index component analysis is performed on the downward correlation index sequence to construct the real economy index prediction model, which specifically includes the following steps:

Step 402, performing standardization processing on all downlink correlation index sequences respectively to obtain a data matrix.

Collect p-dimensional random vector x=(X ₁ ,X ₂ ,...,X _p ) ^T for n downlink correlation index sequences of the same time frequency, each downlink correlation index sequence x _i =(x _i1 ,x _i2 ,...,x _ip ) ^T , i=1,2,…,n, n>p, construct a matrix, and perform the following standardized transformation on the matrix elements to eliminate the dimensional differences and orders of magnitude differences between different index data difference:

in, Get the standardized data matrix Z. n is the total number of downlink correlation indicator sequences, and p is the number of values in the sequence corresponding to each downlink correlation indicator sequence. Z _i represents the data matrix of the i-th downward correlation index sequence, and Z _ij represents the data matrix of the j-th numerical value in the i-th downward correlation index sequence.

Step 404: Obtain the covariance matrix of the downlink correlation index sequence according to the data matrix, and calculate the eigenvalues, eigenvectors, and variance contribution rates of the principal components of the covariance matrix.

According to the data matrix Z, the covariance matrix R is obtained, and the calculation formula of the covariance matrix R is as follows:

Then solve the covariance matrix R to obtain p characteristic roots, and obtain the principal component variance contribution rate of p characteristic roots and the corresponding eigenvectors Convert the standardized indicator variables into principal components according to the characteristic roots and characteristic vectors, as follows:

U ₁ is called the first principal component expression, U ₂ is called the second principal component expression, ..., U _p is called the pth principal component expression.

Step 406, acquiring a preset minimum variance contribution rate threshold.

The server obtains a preset minimum variance contribution rate threshold from the database. When the variance contribution rate is lower than the preset minimum variance contribution rate threshold, the principal component expression has little influence on the construction of the entire real economy index prediction model, and the final result changes are also relatively small, so the minimum variance contribution rate threshold is at It is a threshold parameter designed to reduce the amount of calculation while ensuring the accuracy of the final real economy index forecasting model. The threshold of the minimum variance contribution rate is set by the user according to the current demand, and generally ranges from 0.01% to 20%.

Step 408 , screening out principal component expressions whose variance contribution rate of the principal components is not less than a preset minimum variance contribution rate threshold.

The server compares the principal component variance contribution rates of all principal component expressions with a preset minimum variance contribution rate threshold. When the comparison determines that the principal component variance contribution rate of the principal component expression is not less than (greater than or equal to) the preset minimum variance contribution rate threshold, the server extracts the principal component expression.

Step 410, constructing a real economy index prediction model according to the filtered principal component expressions.

The weighted summation of the screened principal component expressions is obtained to obtain the real economy index prediction model. The dependent variable in the model is the quarterly growth rate of GDP at the current price, and the independent variable is a number of downward correlation index sequences. Each downward correlation index sequence is given different weights, and the weight is the variance contribution rate of each principal component expression.

In the above information forecasting method, when the variance contribution rate is lower than the preset minimum variance contribution rate threshold, the principal component expression has little influence on the construction of the entire real economy index forecasting model, and the final result changes are also relatively small, so The principal component expression corresponding to the minimum variance contribution rate threshold lower than the preset value is deleted. While ensuring the accuracy of the final real economy index prediction model, it further realizes the dimensionality reduction of the downward correlation index sequence and reduces the amount of calculation.

In one embodiment, as shown in Figure 5, constructing real economy index prediction model according to the principal component expression screened out, including:

Step 502, acquiring a preset minimum principal component contribution rate threshold.

The server obtains the preset minimum principal component contribution rate threshold. The preset minimum principal component contribution rate threshold A is to ensure that the information represented by all basic index sequences is effectively used, and the general range can be 70% to 100%.

Step 504 , comparing the sum of the variance contribution rates of all the screened principal components with the preset minimum principal component contribution rate threshold.

The server calculates the sum of the principal component variance contribution rates of all the filtered principal component expressions according to the following formula, and compares it with the preset minimum principal component contribution rate threshold A. At this time, the preset utilization rate A may be set as 85%.

m is the number of final principal component expressions, and λ _j is the principal component variance contribution rate of the jth principal component expression.

Step 506, when the sum of the principal component variance contribution rates is less than the preset minimum principal component contribution rate threshold, calculate a second difference between the two.

When the sum of the principal component variance contribution rates is less than the preset minimum principal component contribution rate threshold, the server calculates a second difference between the principal component variance contribution sum and the preset minimum principal component contribution rate threshold.

Step 508, sort the principal component variance contribution rates of the remaining principal component expressions from large to small, extract the top principal component expressions and the sum of the principal component variance contribution rates of the extracted principal component expressions is not less than second difference.

The server sorts all the principal component expressions from large to small according to the formula in step 504 and the variance contribution rate of the principal components, and gradually extracts the top principal component expressions according to the difference, until all principal component expressions are extracted twice The sum of the variance contribution rates of the principal components of the formula is not less than the second difference. The sum of variance contribution rates of all principal components finally proposed is greater than the preset minimum principal component contribution rate threshold.

This method can construct a multi-period real economy index prediction model in the future, and predict the growth rate of the real economy (current price GDP quarterly) in multiple future periods. When constructing the multi-period real economic index forecasting model in the future, when constructing the model of the next period, the data in the basic index sequence will be moved forward by one, and the original first digit will be discarded. For example, it is possible to set and predict the index data of the next four periods, and extract the four-period basic index series from each basic index data.

As shown in Figure 6, the server adopts this method to construct the real economy index forecasting models for the first period to the fourth period (the second quarter of 2018 to the first quarter of 2019). The initial time of the foreign exchange preset sequence is the second quarter of 2011. The server extracts the growth rate sequence corresponding to the second period (second quarter) of 2011 (the OBJ curve in the figure) from the growth rate value, and the server extracts the growth rate sequence from the basic index data Extract the underlying indicator series corresponding to the initial time of the preset series.

When building the real economy index forecasting model for the second period of 2018, the server extracts the basic index sequence from the second quarter of 2011 to the corresponding first quarter of 2018 from the basic index data. At this time, the real economy index forecasting model It is the OBJ1 curve in the figure. When building the real economy index forecasting model for the third period of 2018, the server extracts the basic index sequence from the second quarter of 2011 to the corresponding first quarter of 2018 from the basic index data, but at this time the server discards the 2011 The value of the foreign exchange growth rate in the second quarter of 2019 and the basic index sequence, so the forecast model of the real economy index is the OBJ2 curve in the figure. When constructing the real economy index forecasting model for the 4th period of 2018, the server extracts the basic index sequence from the 2nd quarter of 2011 to the corresponding 1st quarter of 2018 from the basic index data, and discards the 2011 2nd quarter at the same time. Quarterly and the third quarter of 2011 foreign exchange growth rate value and basic index sequence, so the forecast model of the real economy index is the OBJ3 curve in the figure. When constructing the real economy index forecasting model for the first period of 2019, the server extracts the basic index sequence from the second quarter of 2011 to the corresponding first quarter of 2018 from the basic index data, and discards the second quarter of 2011 at the same time. Quarterly, the third quarter of 2011 and the fourth quarter of 2011 the foreign exchange growth rate value and the basic index sequence, so the forecast model of the real economy index is the OBJ4 curve in the figure.

When using this method to predict the real economic index in different periods, it is necessary to construct the real economic index forecasting model first to ensure the uniqueness of the constructed index in each period. This can reduce the impact caused by changes in policies, situations, and time, and further improve the accuracy of each foreign exchange construction index.

It should be understood that although the various steps in the flow charts of FIGS. 2-5 are shown sequentially as indicated by the arrows, these steps are not necessarily executed sequentially in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least some of the steps in Figures 2-5 may include a plurality of sub-steps or stages, these sub-steps or stages are not necessarily executed at the same time, but may be executed at different times, these sub-steps or stages The order of execution is not necessarily performed sequentially, but may be performed alternately or alternately with at least a part of other steps or sub-steps or stages of other steps.

In one embodiment, as shown in FIG. 7 , an information prediction device is provided, including: a value acquisition module 702, a sequence extraction module 704, a calculation module 706, a screening module 708, a model construction module 710 and an index prediction module 712, in:

The value acquisition module 702 is used to obtain the quarterly growth rate value of GDP at the current price, and obtain the basic index data corresponding to each basic index one by one from the basic index database.

The sequence extraction module 704 is configured to extract multiple time-frequency basic index sequences from each basic index data to obtain multiple sets of basic index sequences.

Calculation module 706, configured to calculate the correlation coefficient between each group of basic index sequences and the growth rate value.

The screening module 708, according to the correlation coefficient, screens out each group of basic index sequences associated with the growth rate value as the downward correlation index sequence.

The model construction module 710 is used to analyze the index components of each group of downward correlation index sequences and construct multiple real economy index prediction models.

The index prediction module 712 is configured to predict the real economy index of the real economy according to multiple real economy index prediction models.

In one embodiment, the calculation module 706 can also be used to substitute the basic index sequence and growth rate values into the correlation calculation formula for calculation; and set the absolute value of the result calculated according to the correlation calculation formula as the correlation coefficient.

In some embodiments, the screening module 708 may also include an extraction unit, a classification unit, and a setting unit, wherein:

The extraction unit is configured to obtain a preset correlation threshold, and extract a basic index sequence with a correlation coefficient greater than the preset correlation threshold as an index sequence to be processed.

The classification unit is used to obtain attribute information corresponding to the index sequence to be processed, and classify the index sequence to be processed according to the attribute information.

The setting unit is used to set the basic index sequence with the largest correlation coefficient in each type of index sequence to be processed as the downward correlation index sequence.

In another embodiment, the screening module 708 may also include an extraction unit, a classification unit, a setting unit, a value acquisition unit, a comparison and judgment unit, and a calculation unit, wherein:

The extraction unit is configured to obtain a preset correlation threshold, and extract a basic index sequence with a correlation coefficient greater than the preset correlation threshold as an index sequence to be processed.

The classification unit is used to obtain attribute information corresponding to the index sequence to be processed, and classify the index sequence to be processed according to the attribute information.

The setting unit is used to set the basic index sequence with the largest correlation coefficient in each type of index sequence to be processed as the downward correlation index sequence.

The value acquisition unit is used to acquire the classification quantity of the index sequence to be processed and the lower limit threshold of the preset index quantity.

The comparison and judgment unit is used for comparing and judging the number of classifications with the lower limit threshold of the number of preset indicators.

When the number of categories is less than the preset lower limit threshold of the number of indicators, the calculation unit is configured to calculate a first difference between the lower limit threshold of the preset number of indicators and the number of categories.

The setting unit is used for sorting the correlation coefficients of the remaining index sequences to be processed from large to small, extracting and setting the index sequences to be processed at the top as downlink correlation index sequences, and the extracted index sequences are at the top The number of pending indicators of is consistent with the first difference.

In another embodiment, the model construction module 710 includes a processing unit, a variance calculation unit, a variance contribution rate acquisition unit, an expression screening unit, and a model construction unit, wherein:

The processing unit is configured to perform standardization processing on all downlink related index sequences to obtain a data matrix.

The variance calculation unit is used to obtain the covariance matrix of the downlink correlation index sequence according to the data matrix, and calculate the characteristic root, characteristic vector and principal component variance contribution rate of the covariance matrix.

The variance contribution rate acquisition unit is configured to acquire a preset minimum variance contribution rate threshold.

The expression screening unit is used to filter out principal component expressions whose variance contribution rate of the principal component is not less than a preset minimum variance contribution rate threshold.

The model construction unit is used for constructing a real economy index prediction model according to the filtered principal component expressions.

In one embodiment, the model construction module 710 includes a processing unit, a variance calculation unit, a variance contribution rate acquisition unit, an expression screening unit, a principal component contribution rate acquisition unit, a judgment unit, a difference calculation unit, and a model construction unit, wherein:

The processing unit is configured to perform standardization processing on all downlink related index sequences to obtain a data matrix.

The variance calculation unit is used to obtain the covariance matrix of the downlink correlation index sequence according to the data matrix, and calculate the characteristic root, characteristic vector and principal component variance contribution rate of the covariance matrix.

The variance contribution rate acquisition unit is configured to acquire a preset minimum variance contribution rate threshold.

The expression screening unit is used to filter out principal component expressions whose variance contribution rate of the principal component is not less than a preset minimum variance contribution rate threshold.

The principal component contribution rate acquisition unit is configured to acquire a preset minimum principal component contribution rate threshold.

The judging unit is configured to compare the sum of the variance contribution rates of all the screened principal components with a preset minimum principal component contribution rate threshold.

The difference calculation unit is configured to calculate a second difference between the two when the sum of the variance contribution rates of the principal components is less than a preset minimum threshold of the principal component contribution rates.

The expression screening unit is also used to sort the principal component variance contribution rates of the remaining principal component expressions from large to small, filter and extract the top principal component expressions and the sum of the extracted principal component variance contribution rates is not less than the second difference.

The model construction unit is used for constructing a real economy index prediction model according to the filtered principal component expressions.

For specific limitations on the information prediction apparatus, refer to the above-mentioned limitations on the information prediction method, which will not be repeated here. Each module in the above-mentioned information prediction device can be fully or partially realized by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, and can also be stored in the memory of the computer device in the form of software, so that the processor can invoke and execute the corresponding operations of the above-mentioned modules.

In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure may be as shown in FIG. 8 . The computer device includes a processor, memory, network interface and database connected by a system bus. Wherein, the processor of the computer device is used to provide calculation and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer programs and databases. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database (basic indicator library) of the computer equipment is used to store the forecast data of the real economic index, the preset relevant threshold, the preset lower limit threshold of the number of indicators, the preset minimum variance contribution rate threshold, the preset minimum principal component contribution rate threshold, the preset The initial time of the sequence, the basic indicator data and its attribute information, etc. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by a processor, a real economy index forecasting method is realized.

Those skilled in the art can understand that the structure shown in FIG. 8 is only a block diagram of a partial structure related to the solution of this application, and does not constitute a limitation on the computer equipment to which the solution of this application is applied. The specific computer equipment can be More or fewer components than shown in the figures may be included, or some components may be combined, or have a different arrangement of components.

In one embodiment, a computer device is provided, including a memory and a processor, the memory is stored with a computer program, and the processor implements the following steps when executing the computer program: obtain the quarterly growth rate value of GDP at the current price, and obtain the quarterly growth rate value from the basic index Obtain the basic index data corresponding to each basic index one by one in the database; extract multiple time-frequency basic index sequences from each basic index data to obtain multiple sets of basic index sequences; calculate the relationship between each set of basic index sequences and the growth rate value Correlation coefficient; according to the correlation coefficient, select the basic index sequence associated with the growth rate value of each group as the downward correlation index sequence; analyze the index components of each group of downward correlation index sequence and construct multiple real economy index prediction models; according to multiple The real economy index forecasting model predicts the real economy index for the real economy.

In one embodiment, when the processor executes the computer program to realize the step of calculating the correlation coefficient between each group of basic index sequence and the growth rate value, it is also used for: substituting the basic index sequence and the growth rate value into the correlation calculation formula. Calculation; set the absolute value of the result calculated according to the correlation calculation formula as the correlation coefficient.

In one embodiment, when the processor executes the computer program to realize the step of screening out the basic index sequence associated with the real economic data as the downward correlation index sequence according to the correlation coefficient, it is also used to: obtain a preset correlation threshold, set the correlation coefficient greater than the preset The basic index sequence of the correlation threshold is extracted as the index sequence to be processed; the attribute information corresponding to the index sequence to be processed is obtained, and the index sequence to be processed is classified according to the attribute information; the basis with the largest correlation coefficient in each type of index sequence to be processed is The index sequence is set as the downward correlation index sequence.

In one embodiment, when the processor executes the computer program to implement the step of setting the basic index sequence with the largest correlation coefficient in each classified index sequence to be processed as the downward correlation index sequence, it is also used to: obtain the classification of the index sequence to be processed Quantity and the lower limit threshold of the number of preset indicators; compare the number of categories with the lower limit threshold of the number of preset indicators; when the number of categories is less than the lower limit threshold of the number of preset indicators, calculate the first difference between the lower limit threshold of the number of preset indicators and the number of categories; The correlation coefficients of the remaining index sequences to be processed are sorted from large to small, and the top index sequences to be processed are also extracted as downlink correlation index sequences, and the number of the extracted index sequences to be processed is the same as that of the first index sequence. A difference is the same.

In one embodiment, when the processor executes the computer program, it implements the step of analyzing the index components of the downward correlation index sequence and constructing the real economy index forecasting model, which is also used to: standardize all the downward correlation index sequences to obtain the data matrix; According to the data matrix, the covariance matrix of the downward correlation index sequence is obtained, and the characteristic root, characteristic vector and principal component variance contribution rate of the covariance matrix are calculated; the preset minimum variance contribution rate threshold is obtained; the variance contribution rate of the principal components is not less than The principal component expression of the preset minimum variance contribution rate threshold; the real economic index prediction model is constructed according to the screened principal component expression.

In one embodiment, when the processor executes the computer program to implement the step of constructing the real economic index prediction model according to the filtered principal component expressions, it is also used to: obtain the preset minimum principal component contribution rate threshold; The sum of the component variance contribution rates is compared with the preset minimum principal component contribution rate threshold; when the sum of the principal component variance contribution rates is less than the preset minimum principal component contribution rate threshold, the second difference between the two is calculated; the remaining The principal component variance contribution rates of the principal component expressions are sorted from large to small, and the top principal component expressions are extracted and the sum of the extracted principal component variance contribution rates is not less than the second difference.

In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

Obtain the quarterly growth rate value of GDP at the current price, and obtain the basic index data corresponding to each basic index one by one from the basic index database;

Extract multiple time-frequency basic index sequences from each basic index data to obtain multiple sets of basic index sequences;

Calculate the correlation coefficient between each group of basic index series and the growth rate value;

According to the correlation coefficient, the basic index sequence associated with the growth rate value of each group is selected as the downward correlation index sequence;

Analyze the index components of each group of downward correlation index sequences and construct multiple real economy index prediction models;

According to multiple real economy index forecasting models, the real economy index is forecasted for the real economy.

In one embodiment, when the computer program is executed by the processor to realize the step of calculating the correlation coefficient between each group of basic index sequence and the growth rate value, it is also used for: substituting the basic index sequence and the growth rate value into the correlation calculation formula Perform calculation; set the absolute value of the result calculated according to the correlation calculation formula as the correlation coefficient.

In one embodiment, when the computer program is executed by the processor to realize the step of screening out the basic index sequence associated with the real economic data as the downward correlation index sequence according to the correlation coefficient, it is also used to: obtain a preset correlation threshold, set the correlation coefficient greater than The basic index sequence with preset correlation threshold is extracted as the index sequence to be processed; the attribute information corresponding to the index sequence to be processed is obtained, and the index sequence to be processed is classified according to the attribute information; the index sequence with the largest correlation coefficient in each type of index sequence to be processed is The basic index sequence is set as the downward correlation index sequence.

In one embodiment, when the computer program is executed by the processor to implement the step of setting the basic index sequence with the largest correlation coefficient in each classified index sequence to be processed as the downward correlation index sequence, it is also used to: obtain the index sequence to be processed The number of categories and the lower threshold of the number of preset indicators; compare the number of categories with the lower threshold of the number of preset indicators; when the number of categories is less than the lower threshold of the number of preset indicators, calculate the first difference between the lower threshold of the number of preset indicators and the number of categories ; Sort the correlation coefficients of the remaining index sequences to be processed from large to small, and extract the top index sequences to be processed as downlink correlation index sequences, and the number of extracted index sequences to be processed is equal to The first difference is the same.

In one embodiment, when the computer program is executed by the processor to implement the step of analyzing the index components of the downward correlation index sequence and constructing the real economic index forecasting model, it is also used to: perform standardization processing on all the downward correlation index sequences to obtain the data matrix ;According to the data matrix, the covariance matrix of the downlink correlation index sequence is obtained, and the characteristic root, characteristic vector and principal component variance contribution rate of the covariance matrix are calculated; the preset minimum variance contribution rate threshold is obtained; the variance contribution rate of the principal components is screened out. A principal component expression that is less than the preset minimum variance contribution rate threshold; construct a real economic index prediction model based on the screened principal component expression.

In one embodiment, when the computer program is executed by the processor to realize the step of constructing the real economic index prediction model according to the filtered principal component expressions, it is also used to: obtain the preset minimum principal component contribution rate threshold; The sum of the principal component variance contribution rate is compared with the preset minimum principal component contribution rate threshold; when the principal component variance contribution rate sum is less than the preset minimum principal component contribution rate threshold, the second difference between the two is calculated; the remaining The principal component variance contribution rates of the principal component expressions are sorted from large to small, and the top principal component expressions are extracted and the sum of the extracted principal component variance contribution rates is not less than the second difference.

Those of ordinary skill in the art can understand that realizing all or part of the processes in the methods of the above embodiments can be completed by instructing related hardware through computer programs, and the computer programs can be stored in a non-volatile computer-readable storage medium , when the computer program is executed, it may include the procedures of the embodiments of the above-mentioned methods. Wherein, any references to memory, storage, database or other media used in the various embodiments provided in the present application may include non-volatile and/or volatile memory. Nonvolatile memory can include read only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Chain Synchlink DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

The technical features of the above embodiments can be combined arbitrarily. To make the description concise, all possible combinations of the technical features in the above embodiments are not described. However, as long as there is no contradiction in the combination of these technical features, they should be It is considered to be within the range described in this specification.

The above-mentioned embodiments only represent several implementation modes of the present application, and the description thereof is relatively specific and detailed, but it should not be construed as limiting the scope of the patent for the invention. It should be noted that those skilled in the art can make several modifications and improvements without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the scope of protection of the patent application should be based on the appended claims.

Claims (10)

1. A method of information prediction, the method comprising:

acquiring a current price GDP quarterly increase rate value, and acquiring basic index data corresponding to each basic index one by one from a basic index library;

extracting a plurality of time-frequency basic index sequences from each basic index data to obtain a plurality of groups of basic index sequences;

calculating a correlation coefficient between each group of the basic index sequence and the growth rate value respectively;

screening out basic index sequences which are associated with the growth rate numerical values and are in each group according to the correlation coefficients to serve as downlink association index sequences;

respectively carrying out index component analysis on each group of downlink associated index sequences and constructing to obtain a plurality of entity economic index prediction models;

and carrying out entity economic index prediction on entity economy according to the entity economic index prediction models.

2. The method according to claim 1, wherein said calculating a correlation coefficient between each set of the base index sequences and the growth rate value comprises:

substituting the basic index sequence and the growth rate numerical value into a correlation calculation formula for calculation;

and setting the absolute value of the result obtained by calculation according to the correlation calculation formula as a correlation coefficient.

3. The method according to claim 1, wherein the screening out, as a downlink association index sequence, each group of basic index sequences associated with the growth rate value according to the correlation coefficients includes:

acquiring a preset correlation threshold, and extracting a basic index sequence with the correlation number larger than the preset correlation threshold as an index sequence to be processed;

acquiring attribute information corresponding to the index sequence to be processed, and classifying the index sequence to be processed according to the attribute information;

and setting the basic index sequence with the maximum relation number in the index sequences to be processed in each classification as a downlink associated index sequence.

4. The method according to claim 3, wherein the step of setting the base index sequence with the largest number of relationships in each classification of the index sequences to be processed as a downlink associated index sequence further comprises:

acquiring the classification number of the index sequence to be processed and a preset index number lower limit threshold;

comparing the classification quantity with the lower limit threshold of the preset index quantity;

when the classification number is smaller than the preset index number lower limit threshold, calculating a first difference value between the preset index number lower limit threshold and the classification number;

and sequencing the correlation coefficients of the remaining index sequences to be processed from large to small, extracting the index sequences to be processed arranged in the front row as downlink correlation index sequences, wherein the number of the extracted index sequences to be processed arranged in the front row is consistent with the first difference.

5. The method according to claim 1, wherein the index component analysis of the downlink correlation index sequence to construct an entity economic index prediction model comprises:

respectively carrying out standardization processing on all the downlink correlation index sequences to obtain a data matrix;

obtaining a covariance matrix of the downlink correlation index sequence according to the data matrix, and calculating to obtain a characteristic root, a characteristic vector and a principal component variance contribution rate of the covariance matrix;

acquiring a preset minimum variance contribution rate threshold;

screening out principal component expressions of which the principal component variance contribution rate is not less than the preset minimum variance contribution rate threshold;

and constructing an entity economic index prediction model according to the screened principal component expression.

6. The method of claim 5, wherein constructing an entity economic index prediction model according to the screened principal component expressions comprises:

acquiring a preset lowest principal component contribution rate threshold;

comparing the sum of all screened principal component variance contribution rates with the preset lowest principal component contribution rate threshold;

when the sum of the principal component variance contribution rates is smaller than the preset lowest principal component contribution rate threshold value, calculating a second difference value between the two;

and sorting the principal component variance contribution rates of the remaining principal component expressions from large to small, extracting the principal component expressions ranked in the front, and the sum of the extracted principal component variance contribution rates being not less than the second difference.

7. An information prediction apparatus, characterized in that the apparatus comprises:

the numerical value acquisition module is used for acquiring the seasonal increase rate numerical value of the current GDP and acquiring basic index data corresponding to each basic index one by one from the basic index library;

the sequence extraction module is used for extracting a plurality of time-frequency basic index sequences from each basic index data to obtain a plurality of groups of basic index sequences;

the calculation module is used for calculating a correlation coefficient between each group of the basic index sequences and the growth rate value respectively;

the screening module screens out each group of basic index sequences associated with the growth rate numerical value according to the correlation coefficient to serve as downlink associated index sequences;

the model construction module is used for carrying out index component analysis on each group of downlink associated index sequences and constructing a plurality of entity economic index prediction models;

and the index prediction module is used for predicting the entity economic index of the entity economy according to the entity economic index prediction models.

8. The apparatus of claim 7, wherein the screening module comprises:

the extraction unit is used for acquiring a preset correlation threshold value and extracting a basic index sequence with the correlation number larger than the preset correlation threshold value as an index sequence to be processed;

the classification unit is used for acquiring attribute information corresponding to the index sequence to be processed and classifying the index sequence to be processed according to the attribute information;

and the setting unit is used for setting the basic index sequence with the maximum relation number in each type of index sequences to be processed as the downlink associated index sequence.

9. A computer device comprising a memory and a processor, the memory storing a computer program, wherein the processor implements the steps of the method of any one of claims 1 to 6 when executing the computer program.

10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 6.