CN117993729A - Method and system for processing public accumulation fund data - Google Patents
Method and system for processing public accumulation fund data Download PDFInfo
- Publication number
- CN117993729A CN117993729A CN202311842058.6A CN202311842058A CN117993729A CN 117993729 A CN117993729 A CN 117993729A CN 202311842058 A CN202311842058 A CN 202311842058A CN 117993729 A CN117993729 A CN 117993729A
- Authority
- CN
- China
- Prior art keywords
- index
- data
- accumulation
- economic
- public
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000009825 accumulation Methods 0.000 title claims abstract description 274
- 238000000034 method Methods 0.000 title claims abstract description 69
- 238000012545 processing Methods 0.000 title claims abstract description 37
- 238000013528 artificial neural network Methods 0.000 claims abstract description 11
- 125000004122 cyclic group Chemical group 0.000 claims abstract description 7
- 239000011159 matrix material Substances 0.000 claims description 46
- 239000013598 vector Substances 0.000 claims description 34
- PCHJSUWPFVWCPO-UHFFFAOYSA-N gold Chemical compound [Au] PCHJSUWPFVWCPO-UHFFFAOYSA-N 0.000 claims description 30
- 239000010931 gold Substances 0.000 claims description 30
- 229910052737 gold Inorganic materials 0.000 claims description 30
- 230000006870 function Effects 0.000 claims description 28
- 238000009826 distribution Methods 0.000 claims description 24
- 230000002195 synergetic effect Effects 0.000 claims description 16
- 238000003860 storage Methods 0.000 claims description 14
- 238000004519 manufacturing process Methods 0.000 claims description 12
- 238000004590 computer program Methods 0.000 claims description 7
- 238000001744 unit root test Methods 0.000 claims description 6
- 238000007781 pre-processing Methods 0.000 claims description 5
- 238000001514 detection method Methods 0.000 claims description 4
- 230000001537 neural effect Effects 0.000 claims description 2
- 238000005457 optimization Methods 0.000 abstract description 16
- 238000012360 testing method Methods 0.000 description 11
- 238000004364 calculation method Methods 0.000 description 9
- 230000000875 corresponding effect Effects 0.000 description 8
- 230000008569 process Effects 0.000 description 7
- 238000003672 processing method Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 5
- 238000007726 management method Methods 0.000 description 5
- 238000013488 ordinary least square regression Methods 0.000 description 5
- 239000000470 constituent Substances 0.000 description 4
- 238000006243 chemical reaction Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 230000002596 correlated effect Effects 0.000 description 2
- 238000013480 data collection Methods 0.000 description 2
- 238000007418 data mining Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 239000004973 liquid crystal related substance Substances 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 238000012216 screening Methods 0.000 description 2
- 230000003321 amplification Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 230000001276 controlling effect Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000000802 evaporation-induced self-assembly Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003199 nucleic acid amplification method Methods 0.000 description 1
- 238000000611 regression analysis Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0637—Strategic management or analysis, e.g. setting a goal or target of an organisation; Planning actions based on goals; Analysis or evaluation of effectiveness of goals
- G06Q10/06375—Prediction of business process outcome or impact based on a proposed change
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Educational Administration (AREA)
- Development Economics (AREA)
- Operations Research (AREA)
- Marketing (AREA)
- Game Theory and Decision Science (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a method and a system for processing public accumulation data, which belong to the technical field of data processing, wherein, firstly, a historical public accumulation index is obtained based on macroscopic economic data and public accumulation data, then, future public accumulation index prediction is carried out through a cyclic neural network based on the historical public accumulation index, and further, macroscopic economic trend is predicted according to the public accumulation index; according to the invention, the weight of the accumulation index is adjusted by considering the stability of the time sequence and the linear optimization of the unconstrained condition, so that the obtained result of predicting the accumulation index is more robust, and the macroscopic economic trend can be predicted better.
Description
Technical Field
The invention belongs to the technical field of data processing, and particularly relates to a method, a system, electronic equipment and a storage medium for processing public accumulation fund data.
Background
The economic index is an important tool for measuring the overall economic activity level of a country or region, and the public accumulation data has strong correlation with the economic index; in recent years, machine learning models are increasingly widely used for processing economic data to meet the demands of people for economic data analysis, economic data mining and economic prediction.
At present, how to utilize a machine learning model to conduct economic index analysis based on the accumulation fund data becomes a problem to be solved.
Disclosure of Invention
The invention provides an accumulation fund data processing method, an accumulation fund data processing system, electronic equipment and a storage medium, which are used for overcoming at least one technical problem in the prior art.
In order to achieve the above object, the present invention provides a method for processing public accumulation fund data, the method comprising: acquiring macroscopic economic data and public accumulation fund data;
constructing a macroscopic economic index time sequence according to macroscopic economic data and acquiring an initial weight matrix of macroscopic economic indexes, constructing an public accumulation index time sequence according to the public accumulation data and acquiring an initial weight matrix of public accumulation indexes;
Determining an initial weight vector according to the initial weight of the macro economic index and an initial weight matrix of the public accumulation index, constructing a coordination model based on the initial weight vector, the macro economic index time sequence and the public accumulation index time sequence, and acquiring the optimal public accumulation index weight based on the coordination model; determining an accumulation index based on the optimal accumulation index weight;
And inputting the time sequence of the accumulation index into a preset circulating neural network for accumulation index prediction and carrying out accumulation index prediction.
Further, the preferred method is that the index of the public accumulation fund data comprises a first order difference of the collection unit quantity, a unit collection proportion, a unit collection amount, a collection base number, amount paid in deposit amount, a collection balance, an account opening unit quantity, a collection unit quantity completed in the current month, a sales state unit quantity, a hold over state unit quantity and a normal state unit quantity;
The index of the macro economic data comprises an enterprise scenic index, an enterprise home confidence index, a domestic total production value, a manufacturing purchasing manager index, a non-manufacturing purchasing manager index, an enterprise number and enterprise profit.
Further, a preferred method of obtaining an initial weight matrix for the metric of the accumulation fund comprises,
Classifying indexes of the public accumulation fund data according to the correlation between the public accumulation fund indexes and the enterprise operating conditions and determining the corresponding classification level;
assigning initial weights to indexes of the public accumulation gold data after the classification and the level corresponding to the classification are determined, and taking Dirichlet distribution as prior distribution of public accumulation gold index prediction to obtain an initial weight matrix of the public accumulation gold indexes; wherein, the probability density function of dirichlet distribution is:
Wherein, X is a multi-element vector group,Alpha is the parameter set and is used for the control of the system,Α determines the peak and area of the distribution x i, and as α is greater, the distribution of x i on the coordinate axis becomes more and more concentrated in the central area; and the larger alpha i, the larger the corresponding generated x i.
Further, the preferred method further comprises performing a stationarity test on the principal moment index data and the macro economic index data before obtaining the optimal principal moment index vector based on the coordination model, the method comprising,
Constructing a regression equation between the real index data of the public accumulation fund and the predictive index data of the public accumulation fund by using the OLS;
Carrying out unit root test on the macro economic index time sequence and the accumulation gold index time sequence by utilizing the regression equation;
when it is determined that there is no unit root, a synergistic relationship exists between the macro economic indicator time series and the public accumulation gold indicator time series.
Further, the preferred method is that the unit root test of the macro economic index time series and the accumulation gold index time series by using the regression equation is realized by the following formula:
A unit root exists in H 0：b1 =0, and a synergistic relationship does not exist between the macro economic index time series and the public accumulation gold index time series;
H 1：b1 is not equal to 0, no unit root exists, and a synergistic relationship exists between the macro economic index time sequence and the accumulation gold index time sequence;
Wherein, E t＝b0+b1*∈t-1;∈t is the residual of the regression equation; h 0 is that there is at least one unit root in the time sequence; h 1 is that one unit root does not exist for the hypothetical time series.
Further, the method preferably further comprises performing data dimensionalization preprocessing on the accumulation data by the following formula before constructing an accumulation index time sequence according to the accumulation data and acquiring an initial weight matrix of the accumulation index;
Xscaled＝Xstd*(xmax-xmin)+xmin
Wherein X is a matrix of the public accumulation gold data according to time sequence, X ij is data of the ith row and the jth column, X jmin is a minimum value of the jth column of the matrix, and X jmax is a maximum value of the jth column of the matrix. x max is the maximum of the normalized range; x min is the normalized range minimum.
Further, the method preferably further comprises preprocessing the macro economic data before constructing the macro economic index time sequence according to the macro economic data and acquiring the initial weight matrix of the macro economic index:
Filling the macro economic data, and respectively interpolating data among quarter values of the macro economic data by using a linear difference method to obtain macro economic month data;
And respectively interpolating data between two quarters based on the month data to obtain macroscopic economic quarter data.
In order to solve the above problems, the present invention also provides an integrated deposit data processing system, the system comprising,
The data acquisition unit is used for acquiring macroscopic economic data and public accumulation fund data;
The initial weight acquisition unit is used for constructing a macroscopic economic index time sequence according to macroscopic economic data and acquiring an initial weight matrix of the macroscopic economic index, and constructing a public accumulation index time sequence according to the public accumulation data and acquiring an initial weight matrix of the public accumulation index; determining an initial weight vector according to the initial weight of the macro economic index and an initial weight matrix of the public accumulation index, constructing a coordination model based on the initial weight vector, the macro economic index time sequence and the public accumulation index time sequence, and acquiring the optimal public accumulation index weight based on the coordination model; determining an accumulation index based on the optimal accumulation index weight;
And the prediction unit is used for inputting the time sequence of the accumulation index into a preset circulating neural network for accumulation index prediction and performing accumulation index prediction.
In order to solve the above problems, the present invention also provides an electronic device including: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform steps in an aggregate data handling method as described above.
In order to solve the above-mentioned problems, the present invention also provides a computer-readable storage medium storing a computer program which, when executed by a processor, implements an accumulation fund data processing method as described above.
The method, the system, the electronic equipment and the storage medium for processing the public accumulation fund data have the following beneficial effects: firstly, acquiring historical public accumulation indexes based on macroscopic economic data and public accumulation data, and then predicting future public accumulation indexes through a cyclic neural network based on the historical public accumulation indexes, so as to predict macroscopic economic trend according to the public accumulation indexes; according to the invention, the weight of the accumulation index is adjusted by considering the stability of the time sequence and the linear optimization of the unconstrained condition, so that the obtained result of the accumulation index is more robust, and the macroscopic economic trend can be predicted better.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of an embodiment of a method for processing accumulation fund data;
FIG. 2 is a schematic diagram of an embodiment of a method for processing accumulated gold data;
FIG. 3 is a block diagram of an embodiment of a system for processing an accumulation fund data;
FIG. 4 is a schematic diagram of an internal structure of an electronic device for implementing an embodiment of the present invention;
FIG. 5 is a graph showing a predicted comparison of the accumulation index and the accumulation index according to an embodiment of the present invention.
The achievement of the objects, functional features and advantages of the present invention will be further described with reference to the accompanying drawings, in conjunction with the embodiments.
Detailed Description
It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
FIG. 1 depicts the overall method of processing the aggregate data. FIG. 1 is a flow chart of a method for processing public accumulation fund data according to an embodiment of the present invention; the method may be performed by a system, which may be implemented in software and/or hardware.
As shown in fig. 1, in the present embodiment, the public accumulation fund data processing method includes steps S110 to S140.
S110, acquiring macro economic data and accumulation fund data.
Specifically, the data collection is the collection of the principal related data and the macro economic operation data based on the principal of principal design. And performing the accumulation index screening and the macro index screening.
In the prior art, the calculation method of the public accumulation index is to screen data fields related to macroscopic economy in the public accumulation business, and the fields are the constituent indexes of the public accumulation index. The historical data of different accumulation indexes are given different weights, and the weighted accumulation indexes in each historical period are summed to obtain the total accumulation index. Therefore, the design principle of the accumulation index is mainly the correlation, the calculability and the classification principle. The correlation principle refers to that the index needs of the accumulation index calculation are correlated with enterprise operation, and different initial weights are given to the accumulation index according to the degree of correlation with the enterprise operation. Calculability means that the index used to calculate the accumulation index needs to be numerical data and has calculability in the time dimension. The classification principle refers to enterprises and institutions aiming at different classification properties in the accumulation index. The index selection of the macro-economic index calculation mainly follows the correlation principle, and the macro-economic index has different branches, and the macro-economic branch of the enterprise operation condition is exemplified, so that the index of the selected area is the macro-economic index highly correlated with the enterprise operation.
The indexes of the public accumulation fund data comprise first-order difference of the collection unit quantity, unit collection proportion, unit collection amount, collection base number, amount paid in deposit amount, collection balance, account opening unit quantity, the collection unit quantity completed in the current month, account selling state unit quantity, hold over state unit quantity and normal state unit quantity; the index of the macro economic data comprises an enterprise scenic index, an enterprise home confidence index, a domestic total production value, a manufacturing purchasing manager index, a non-manufacturing purchasing manager index, an enterprise number and enterprise profit. In one specific example, nationwide data and Ji Nanshi data are selected and other nationwide data are fitted according to the ratio between Ji Nanshi GDP and nationwide GDP; for data with different time frequencies, an interpolation method-a bar interpolation method is adopted to change the time frequency.
After the accumulation fund data and the macro economic data are collected according to the indexes, the acquired data need to be preprocessed in order to improve the data quality.
Preprocessing of the aggregate data includes, but is not limited to, data de-dimensionalization. The data dimensionality removal refers to a data mining method that data of different measurement standards are classified into a [0.001,1.001] interval in a normalization mode, and then calculation is carried out.
The formula for data dimensionalization is as follows:
Xscaled＝Xstd*(xmax-xmin)+xmin
Wherein X is a matrix of the public accumulation fund data according to time sequence, X ij is data of the ith row and the jth column, X jmin is a minimum value of the jth column of the matrix, and X jmax is a maximum value of the jth column of the matrix, which is 0.001 in the embodiment. x max is the maximum of the normalized range; x min is the minimum of the normalized range, which in this example is 1.001.
Methods of preprocessing the macro-economic data include, but are not limited to, padding the macro-economic data and time-to-data frequency conversion. Filling the macro economic data, and respectively interpolating data among quarter values of the macro economic data by using a linear difference method to obtain macro economic month data; and respectively interpolating data between two quarters based on the month data to obtain macroscopic economic quarter data.
Specifically, each index data of the macro economy needs to be filled with missing data after taking the number, and the traditional filling method comprises forward filling (ffill), namely filling the last non-empty data in time arrangement at the null value, and backward filling (bfill), namely filling the next non-empty data in time arrangement at the null value. Considering that macroscopic economic data has continuity in time, simple forward filling or backward filling cannot reflect actual economic conditions, so that a filling method for filling 5 non-empty data before and after taking a null value is adopted in the invention. The method for realizing the filling can be Lagrangian filling, interpolation method-bar interpolation and simple linear interpolation method filling (2 non-null values before and after null values respectively). In a specific implementation process, according to actual data testing of an accumulation index, the interpolation method-strip interpolation method results are optimal in most cases. In the example of "Jinan corporation's business overstock index", interpolation-stripe interpolation is adopted for data filling. In practical application, the data can be filled by freely selecting the three methods and simple data filling methods such as front filling and back filling according to the data structure.
Specifically, the time data frequency conversion adopts a linear difference method to respectively perform difference on data among quarter values so as to obtain month data.
The linear interpolation method is to assume that there is a missing value (x i,yi) between two points (x 0,y0) and (x 1,y1), and then the calculation formula of y i is:
In the process of converting quarter data, interpolation can be carried out on data between two quarters respectively, so that the problem of inaccurate data conversion caused by integral interpolation is prevented.
In addition, since some of the macroscopic economic data belongs to nationwide data and has extremely high correlation with GDP, the data will be consistent with the data trend of nationwide GDP. Taking GDP of Ji Nanshi as an example, since the scenic index, confidence index, purchasing manager index and non-manufacturing purchasing manager index are national data, there is no data local to Jinan; therefore, the GDP of Jinan/GDP of the whole country is calculated to be a ratio, named as { g n }; wherein { g n } is a one-dimensional vector; the scenery index denominated b n},{bn is a one-dimensional vector.
The data of the ratio of GDP of ji Nanshi to national GDP over time were fitted with polynomials { g n } and { b n } as time series, assuming that the series satisfies the polynomial function, namely:
Here, { g n } is an independent variable, { b n } is an independent variable, and the values of the constants a, bc, d are calculated.
That is, the ratio { g 1,g2,g3,....,gn } of Ji Nanshi GDP to national GDP is fitted by the above function, the parameter set { a, b, c, d, e } is obtained, then the parameter { a, b, c, d } is fixed, and the new residual e is calculated by substituting the values of a, bc, d obtained by the above formula into the following formula:
The treated indexes are as follows:
In the above formula, the subscript i is any one of a scenic index, a confidence index, a purchasing manager index, and a non-manufacturing purchasing manager index. Namely: the 4-class index is transformed by the same method to obtain a new scenic index, confidence index, purchasing manager index and non-manufacturing purchasing manager index which are strongly related to Jinan city. The 4-class index is one of factors for constructing the macro index, and the 4-class index in each period is weighted according to the initial weight of the macro index and the final macro weight is calculated.
S120, constructing a macroscopic economic index time sequence according to macroscopic economic data, acquiring an initial weight matrix of the macroscopic economic index, constructing an accumulation index time sequence according to the accumulation data, and acquiring the initial weight matrix of the accumulation index.
The method for obtaining the initial weight matrix of the accumulation fund indicator comprises the following steps.
S121, classifying indexes of the public accumulation fund data according to the correlation between the public accumulation fund indexes and the enterprise operating conditions, and determining the corresponding classification level.
Still take the enterprise business accumulation of Ji Nanshi as an example:
According to the correlation between the public accumulation index and the enterprise operation condition, the enterprise operation public accumulation index in Jinan city is divided into 3 major categories: the primary, secondary and tertiary indexes and the classification results are shown in table 1.
Table 1 Ji Nanshi Enterprise operation accumulation index classification table
| Public accumulation fund index name | Accumulation fund index field | Classification level | Classification level field | Correlation direction |
| First order difference of collection unit quantity | hjdwsl | First level | tier_1 | Forward direction |
| Unit paying-out proportion | jcbl | Three stages | tier_3 | Forward direction |
| Unit deposit amount | jcje | First level | tier_1 | Forward direction |
| Payment base | jcjs | Three stages | tier_3 | Forward direction |
| Real payment amount | sjce | First level | tier_1 | Forward direction |
| Paying balance | jcye | First level | tier_1 | Forward direction |
| Number of account opening units | khdwsl | First level | tier_1 | Forward direction |
| The number of collection and payment units is finished in the same month | wchj | First level | tier_1 | Forward direction |
| Number of pin status units | xhzt | Second-level | tier_2 | Negative going |
| Hold over number of state units | hjzt | Second-level | tier_2 | Negative going |
| Number of units in normal state | zczt | Second-level | tier_2 | Forward direction |
The index selection results of the macro economic data are shown in table 2 by taking the enterprise operation height correlation as a selection basis.
TABLE 2 index Table of macroscopic economic data
S122, giving initial weight to the indexes of the accumulation fund data after the classification and the level corresponding to the classification are determined, and taking Dirichlet distribution as prior distribution of accumulation fund index prediction to obtain an initial weight matrix of the accumulation fund indexes. A time series of the initial index of the accumulation fund is formed by the initial weights.
The dirichlet distribution (DIRICHLET DISTRIBUTION) is a posterior distribution of polynomial distributions, by parameter sets:
dirichlet distribution defines a set of multi-element vectors:
the probability density function of dirichlet distribution is:
Wherein, X is a multi-element vector group,Alpha is the parameter set and is used for the control of the system,Α determines the peak and area of the distribution x i, and as α is greater, the distribution of x i on the coordinate axis becomes more and more concentrated in the central area; and the larger alpha i, the larger the corresponding generated x i.
In the experimental process, k is set as the number of groups to be grouped, and X= { X 1,x2,x3 } generated after the equal proportion amplification of alpha is the final initial weight.
Specifically, the dirichlet distribution is taken for generating a vector X of addition sum 1, i.e.:
X= { X 1,x2,x3 }, and
The index for representing three kinds of different accumulation data is used as an initial weight of a constituent factor of the accumulation index in constituting the accumulation index. Wherein the dirichlet parameterFor controlling the initial weight magnitude generated.
In a specific implementation process, the correlation between the principal index constituent factors and the macroscopic economic index can be determined according to manual experience, and a factor major class with high correlation is given higher weight, namely elements in the dirichlet parameter matrix are given larger values, and otherwise, the factors are reversed.
Still take the enterprise business accumulation of Ji Nanshi as an example: the index weight classification of the macroindex is similar to the weight setting of the metric index. The macroindex index is classified into four categories in consideration of correlation with business operations, as shown in table 3.
Table 3 index classification table of macro economic data of Ji Nanshi
| Macroindex index level | Level field | Level weight | Nominal parameters | Magnification factor |
| First level | tier_1 | 0.5 | 100 | 50 |
| Second-level | tier_2 | 0.2 | 100 | 20 |
| Three stages | tier_3 | 0.18 | 100 | 18 |
| Four-stage | tier_4 | 0.12 | 100 | 12 |
The alpha of the index dirichlet parameter for the different levels of the public accumulation fund data is set as shown in table 4 below.
Table 4 index weight table of the public accumulation fund data of enterprise operation public accumulation fund of Ji Nanshi
| Index level of accumulation index | Level field | Level weight | Magnification factor | Magnification factor |
| First level | tier_1 | 0.7 | 100 | 700 |
| Second-level | tier_2 | 0.2 | 100 | 200 |
| Three stages | tier_3 | 0.1 | 100 | 100 |
According to dirichlet distribution (DIRICHLET DISTRIBUTION), multiplying the parameter α by the nominal magnification N (n=100), the number of experiments being 1, to control the regional kurtosis of the initial value distribution of the generated X, and finally obtaining the initial weight matrix vector of the metric of the accumulation gold: { x 1,x2,x3 }. Finally, confirming the initial weight of each index according to the number of the levels as follows:
k is the number of indices contained in the i-th vector.
The index processing results of the macro economic data of different levels are shown in table 5.
TABLE 5 index treatment results for different levels of macroeconomic data
Macroindex calculation a weighted average is calculated from the sum of the index data after weighting.
Wherein:
the weight of each index of the macro index is the dirichlet allocation result of the classification level corresponding to the index divided by the number of all indexes in the classification, namely:
for the i-th level, k is the number of indices included in the i-th level.
The calculation method of the accumulation index is that each index is added after weighting, namely:
idx i is the normalized value of a certain index in the macroscopic index.
Specifically, the obtained accumulation index can be used as a leading indicator for macroscopic economy judgment, and the effect of macroscopic economy judgment is achieved by predicting future accumulation indexes.
S130, determining an initial weight vector according to the initial weight of the macro economic index and an initial weight matrix of the public accumulation index, constructing a coordination model based on the initial weight vector, the macro economic index time sequence and the public accumulation index time sequence, and acquiring the optimal public accumulation index weight based on the coordination model; and determining a public accumulation index based on the optimal public accumulation index weight.
The basic idea of constructing the synergistic function is to fix the initial weight matrix vector { W initial } of the calculated macroscopic index and the metric index, and then assume that the final metric index optimal parameter vector is obtained by adding a certain nominal coefficient vector { W adjust } on the basis of { W initial }. The optimal { W adjust } is calculated by a nonlinear optimization method and finally the optimal metric weight { W optimal } is calculated by a method of adding the initial weight vector and the nominal coefficient vector, and the metric index is calculated by weighting { W optimal }, namely:
W adjust = min (g (X)), g (X) being a non-linearly optimised objective function
Woptimal＝Winitial+Wadjust
The collaborative model is constructed as follows:
f(x)＝coint(macrot,pft)
Wherein macro t is a macroexponent, pf t is an accumulation index, coint is a synergistic function, g (W adjust) =min (f (x)).
In processing two time sequences, if there is a time sequence such that the combination of two non-stationary time sequences is stationary, the two sequences are coordinated, i.e. the two non-stationary time sequences have a long-term stationary relationship within the sample interval. For the present invention, a synergistic model is constructed by constructing the macro economic indicator time series and the aggregate indicator time series.
The method comprises the steps of fixing the weight parameters of macro economic index factors, fixing the initial weight parameters of the accumulation index constituent factors, and constructing a synergistic equation function of a macro economic index time sequence and an accumulation index time sequence on the premise of adjusting the non-fixed weight parameters. And (3) taking the synergistic equation test P value as a function target, taking the public accumulation index adjustment weight parameter as an independent variable, minimizing the P value in a nonlinear optimization mode, and obtaining the self-variable value, namely obtaining the public accumulation index adjustment weight which enables the P value to be minimized, namely the best synergistic effect of the macroscopic economic index and the public accumulation index. That is, the initial value of the principal sketch and the macroscopic economic index are coordinated and nonlinear optimization is performed through a coordinated objective (P value is minimum) to form an index weighted value (i.e. an optimal principal sketch weight), and the initial weight of the principal sketch and the weighted value of the optimal principal sketch weight are added to obtain the final principal sketch.
The basic idea of hypothesis testing is the principle of "small probability events", whose statistical inference method is a countercheck with some probability property. The idea of small probability means that a small probability event does not substantially occur in one trial. The countercheck concept is to put forward a test hypothesis first, then use a proper statistical method, and determine whether the hypothesis is established by using a small probability principle. That is, when the P value of the hypothesis test is a function target, the P value of the hypothesis test, i.e., the probability that the false rejection of the original hypothesis may occur, is calculated, and if the P value is less than the probability threshold value (α) of the small probability time set in advance, the original hypothesis may be rejected on the basis that the confidence interval (confidence interval) is (1- α).
In a specific implementation process, if the P value is not used as a function target in the cooperative equation test, the method can be implemented as follows: hypothesis testing for a certain parameter x refers to putting forward a hypothesis value mu 0 for the x value of the overall, setting an original hypothesis H 0 and a candidate hypothesis H 1, calculating a z statistical value, and comparing the z statistical value with a critical value (critical value) to determine whether the original hypothesis needs to be rejected, if the z value is greater than the critical value (critical value), rejecting the original hypothesis, and if the z value is less than the critical value (critical value), not rejecting the original hypothesis.
Before acquiring the optimal public accumulation index weight based on the coordination model, the method also comprises the step of carrying out stability detection on the public accumulation index data and the macroscopic economic index data, wherein the method comprises the steps of,
S131, constructing a regression equation between the real index data of the accumulation fund and the predictive index data of the accumulation fund by utilizing the OLS, wherein the regression equation is shown in the following formula. The OLS (Ordinary Least Square common least squares) function performs regression analysis.
Yt＝β*Xt+εt
Wherein Y t is a macroscopic economic index time sequence, and X t is the public accumulation gold index time sequence. Beta is the coefficient calculated for the OLS regression between the macroeconomic index and the overstock index.
And S132, carrying out unit root test on the macro economic index time sequence and the accumulation gold index time sequence by utilizing the regression equation.
Performing stationarity detection on residual E t in the regression equation, and firstly, constructing a first-order autoregressive equation E t＝b0+b1*∈t-1 on residual E t; then, ADFuller unit root test is performed on b 1, b 0 is the set intercept, and the requirement for residual stability is b 1 +.1.
H 0：b1 =1, i.e. a unit root exists, and no synergistic relationship exists between the macro economic index time sequence and the accumulation gold index time sequence;
H 1：b1 +.1, i.e. there is no unit root, there is a synergistic relationship between the macro economic indicator time series and the accumulation gold indicator time series.
H 0 is that there is at least one unit root in the time sequence; h 1 is that one unit root does not exist for the hypothetical time series.
S133, when it is judged that the unit root does not exist, a synergistic relationship exists between the macro economic index time sequence and the accumulation gold index time sequence.
Taking the P value returned by the function, namely the probability value of the false rejection original hypothesis of hypothesis test, if the P value is smaller than the probability critical value (alpha), the probability that the event of the macro economic index time sequence Y t and the public accumulation gold index time sequence X t which do not have the cooperative relationship is a small probability event, namely the time sequences Y t and X t are considered to have the cooperative relationship in the confidence interval of (1-alpha).
A nonlinear optimization process for the coordination model is also included.
The unconstrained nonlinear optimization is to calculate the function value which minimizes the objective function value on the basis of the unconstrained function solution, and the mathematical expression is as follows:
fmin(X)＝min(f(X))
In a specific embodiment, a loop calculation method is adopted to select a minimized initial point, where the initial point refers to a point near which the fmin function needs to perform minimum solution on the parameter { X }, where the initial point is different, and the result of the function solution is different. The present embodiment is designed for the parameter x= { X 1,x2,x3 } as follows:
x 1 is a value range (-0.5,0,0.5,1,1.5); x 2 is a value range [0]; x 3 ranges from value 0.
And (3) arranging and combining in the above value range to form X= { X 1,x2,x3 }, carrying out a minimization function solution on each X= { X 1,x2,x3 }, and finally, integrating the P value and the W adjust result of different X= { X 1,x2,x3 }, so as to screen the final X= { X 1,x2,x3 }. It should be noted that X refers to an initial point at which the linear programming is minimized; since there are 3 vectors to be linearly planned, the initial point is a three-dimensional vector. The initial point is selected in relation to the result of the final linear optimization minimum, and therefore, the initial point is selected in a cyclic manner to find the optimal linear optimization result.
That is, the value of the self-variable is obtained by minimizing the value of the P by taking the value of the co-ordination equation test P as a function target and taking the value of the accumulation index adjustment weight parameter as an independent variable and by adopting a nonlinear optimization mode, namely, the accumulation index adjustment weight which enables the value of the P to be minimized, namely, the optimal synergistic effect of the macroscopic economic index and the accumulation index is obtained.
S140, inputting the time sequence of the accumulation index into a preset circulating neural network for accumulation index prediction and carrying out accumulation index prediction. The prediction of the business index of the public accumulation enterprise is based on a cyclic neural grid model under a deep learning framework, and the effect of automatically searching the optimal value of the data prediction of different data structures is achieved through the optimal parameter combination of grid search.
Fig. 2 is a schematic diagram of an embodiment of an accumulation data processing method according to the present invention. Fig. 2 is a schematic diagram of an embodiment of a method for processing public accumulation fund data. As shown, data collection and cleaning are performed first, specifically including collecting macro economic index data highly related to macro economic development and public accumulation business index data related to enterprise operations. And then, respectively carrying out macro economic index and accumulation gold index calculation on the collected and cleaned data, endowing different indexes with initial weight values according to the correlation, and carrying out normalization processing on the initial weight values of the finally obtained indexes.
And finally, constructing a time sequence of macroscopic economic index data and a time sequence of the public accumulation business index data, and checking the stable determination and setting of adjustment factors of the time sequence of the macroscopic economic index data and the time sequence of the public accumulation business index data to perform unconstrained condition linear optimization. Linear optimization initial parameter optimization based on grid search; and calculating the accumulation index according to the adjustment factors and predicting the accumulation index. For example, the actual calculated accumulation index data may only be up to month 6 of 2023, while the predicted accumulation index data may be month 6 of 2024. Model deployment: and deploying the model after the grid search optimization parameters are stabilized, and realizing the display and application of the results.
In a specific embodiment, the accumulation index of 2021, month 7, year 2022, month 11 is predicted by the model of the present invention based on the accumulation index data of 2016, month 1, year 2021, month 6. Then comparing the obtained predicted accumulation index of 2021 month 7 to 2022 month 11 with the actually generated accumulation index of 2021 month 7 to 2022 month 11; the accuracy of the log index prediction of the present invention is shown in FIG. 5.
As can be seen from the observation of FIG. 5, the predictive accuracy of the accumulation index of the accumulation data processing method of the invention reaches more than 70%, and the average accuracy reaches about 93%. Moreover, the macroscopic index trend and the accumulation index trend of fig. 5 clearly show consistency, and thus, macroscopic economic prediction of the macroscopic index can be performed on the basis of the predicted accumulation index data obtained by the present invention.
In summary, the invention obtains the historical public accumulation index based on the macroscopic economic data and the public accumulation data, predicts the future public accumulation index based on the historical public accumulation index through the cyclic neural network, and further predicts the macroscopic economic trend according to the public accumulation index; according to the invention, the weight of the accumulation index is adjusted by considering the stability of the time sequence and the linear optimization of the unconstrained condition, so that the obtained result of the accumulation index is more robust, and the macroscopic economic trend can be predicted better.
As shown in FIG. 3, the present invention provides an integrated deposit data processing system 300 that may be installed in an electronic device. Depending on the functions implemented, the aggregate data processing system 300 may include a data acquisition unit 310, an initial weight acquisition unit 320, and a prediction unit 330. The inventive unit, which may also be referred to as a module, refers to a series of computer program segments, which are stored in the memory of the electronic device, capable of being executed by the processor of the electronic device and of performing a fixed function.
In the present embodiment, the functions concerning the respective modules/units are as follows:
a data acquisition unit 310 for acquiring macro economic data and accumulation fund data;
An initial weight acquisition unit 320, configured to construct a macro economic indicator time sequence according to macro economic data and acquire an initial weight matrix of the macro economic indicator, and construct an accumulation gold indicator time sequence according to the accumulation gold data and acquire an initial weight matrix of the accumulation gold indicator; determining an initial weight vector according to the initial weight of the macro economic index and an initial weight matrix of the public accumulation index, constructing a coordination model based on the initial weight vector, the macro economic index time sequence and the public accumulation index time sequence, and acquiring the optimal public accumulation index weight based on the coordination model; determining an accumulation index based on the optimal accumulation index weight;
And a prediction unit 330, configured to input the time sequence of the accumulation index into a preset circulating neural network for accumulation index prediction and perform accumulation index prediction.
The accumulation data processing system 300 acquires historical accumulation indexes based on macroscopic economic data and accumulation data, predicts future accumulation indexes through a cyclic neural network based on the historical accumulation indexes, and further predicts macroscopic economic trend according to the accumulation indexes; according to the invention, the weight of the accumulation index is adjusted by considering the stability of the time sequence and the linear optimization of the unconstrained condition, so that the obtained result of the accumulation index is more robust, and the macroscopic economic trend can be predicted better.
As shown in fig. 4, the invention also provides an electronic device 4 of the public accumulation fund data processing method.
The electronic device 4 may comprise a processor 40, a memory 41 and a bus, and may further comprise a computer program, such as an accumulation fund data processing program 42, stored in the memory 41 and executable on said processor 40. Memory 41 may also include both internal storage units and external storage devices of the aggregate data handling system. The memory 41 may be used not only for storing codes or the like installed in application software and various types of data such as an accumulation fund data processing program, but also for temporarily storing data that has been output or is to be output.
The memory 41 includes at least one type of readable storage medium, including flash memory, a mobile hard disk, a multimedia card, a card memory (e.g., SD or DX memory, etc.), a magnetic memory, a magnetic disk, an optical disk, etc. The memory 41 may in some embodiments be an internal storage unit of the electronic device 4, such as a removable hard disk of the electronic device 4. The memory 41 may also be an external storage device of the electronic device 4 in other embodiments, such as a plug-in mobile hard disk, a smart memory card (SMART MEDIA CARD, SMC), a Secure Digital (SD) card, a flash memory card (FLASH CARD) or the like, which are provided on the electronic device 4. Further, the memory 41 may also include both an internal storage unit and an external storage device of the electronic device 4. The memory 41 may be used not only for storing application software installed in the electronic device 4 and various types of data, such as codes of an accumulation fund data processing program, but also for temporarily storing data that has been output or is to be output.
The processor 40 may be comprised of integrated circuits in some embodiments, for example, a single packaged integrated circuit, or may be comprised of multiple integrated circuits packaged with the same or different functions, including one or more central processing units (Central Processing unit, CPU), microprocessors, digital processing chips, graphics processors, various control chips, and the like. The processor 40 is a Control Unit (Control Unit) of the electronic device, connects various parts of the entire electronic device using various interfaces and lines, executes programs or modules (e.g., an aggregate data processing program, etc.) stored in the memory 41 by running or executing the programs or modules, and invokes data stored in the memory 41 to perform various functions of the electronic device 4 and process the data.
The bus may be a peripheral component interconnect standard (PERIPHERAL COMPONENT INTERCONNECT, PCI) bus, or an extended industry standard architecture (extended industry standard architecture, EISA) bus, among others. The bus may be classified as an address bus, a data bus, a control bus, etc. The bus is arranged to enable a connection communication between the memory 41 and at least one processor 40 etc.
Fig. 4 shows only an electronic device with components, it will be understood by those skilled in the art that the structure shown in fig. 4 is not limiting of the electronic device 4 and may include fewer or more components than shown, or may combine certain components, or a different arrangement of components.
For example, although not shown, the electronic device 4 may further include a power source (such as a battery) for powering the various components, and the power source may be logically connected to the at least one processor 40 via a power management system, such that functions of charge management, discharge management, and power consumption management are performed by the power management system. The power supply may also include one or more of any of a direct current or alternating current power supply, a recharging system, a power failure detection circuit, a power converter or inverter, a power status indicator, and the like. The electronic device 4 may further include various sensors, bluetooth modules, wi-Fi modules, etc., which will not be described herein.
Further, the electronic device 4 may also comprise a network interface, optionally comprising a wired interface and/or a wireless interface (e.g. WI-FI interface, bluetooth interface, etc.), typically used for establishing a communication connection between the electronic device 4 and other electronic devices.
The electronic device 4 may optionally further comprise a user interface, which may be a Display, an input unit, such as a Keyboard (Keyboard), or a standard wired interface, a wireless interface. Alternatively, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch, or the like. The display may also be referred to as a display screen or display unit, as appropriate, for displaying information processed in the electronic device 4 and for displaying a visual user interface.
It should be understood that the embodiments described are for illustrative purposes only and are not limited to this configuration in the scope of the patent application.
The public accumulation data processing program 42 stored in the memory 41 of the electronic device 4 is a combination of instructions that, when executed in the processor 40, implement: acquiring macroscopic economic data and public accumulation fund data; constructing a macroscopic economic index time sequence according to macroscopic economic data and acquiring an initial weight matrix of macroscopic economic indexes, constructing an public accumulation index time sequence according to the public accumulation data and acquiring an initial weight matrix of public accumulation indexes; determining an initial weight vector according to the initial weight of the macro economic index and an initial weight matrix of the public accumulation index, constructing a coordination model based on the initial weight vector, the macro economic index time sequence and the public accumulation index time sequence, and acquiring the optimal public accumulation index weight based on the coordination model; determining an accumulation index based on the optimal accumulation index weight; and inputting the time sequence of the accumulation index into a preset circulating neural network for accumulation index prediction and carrying out accumulation index prediction.
In particular, the specific implementation method of the above instructions by the processor 40 may refer to the description of the relevant steps in the corresponding embodiment of fig. 1, which is not repeated herein. It should be emphasized that, to further ensure the privacy and security of the public accumulation fund data processing procedure, the database high availability processing data is stored in the node of the blockchain where the server cluster is located.
Further, the integrated modules/units of the electronic device 4 may be stored in a computer readable storage medium if implemented in the form of software functional units and sold or used as separate products. The computer readable medium may include: any entity or system capable of carrying the computer program code, a recording medium, a U disk, a removable hard disk, a magnetic disk, an optical disk, a computer Memory, a Read-Only Memory (ROM).
Embodiments of the present invention also provide a computer readable storage medium, which may be non-volatile or volatile, storing a computer program which when executed by a processor implements: acquiring macroscopic economic data and public accumulation fund data; constructing a macroscopic economic index time sequence according to macroscopic economic data and acquiring an initial weight matrix of macroscopic economic indexes, constructing an public accumulation index time sequence according to the public accumulation data and acquiring an initial weight matrix of public accumulation indexes; determining an initial weight vector according to the initial weight of the macro economic index and an initial weight matrix of the public accumulation index, constructing a coordination model based on the initial weight vector, the macro economic index time sequence and the public accumulation index time sequence, and acquiring the optimal public accumulation index weight based on the coordination model; determining an accumulation index based on the optimal accumulation index weight; and inputting the time sequence of the accumulation index into a preset circulating neural network for accumulation index prediction and carrying out accumulation index prediction.
In particular, the specific implementation method of the computer program when executed by the processor may refer to the description of the relevant steps in the embodiment of the accumulation fund data processing method, which is not repeated herein.
In the several embodiments provided by the present invention, it should be understood that the disclosed apparatus, system and method may be implemented in other manners. For example, the system embodiments described above are merely illustrative, e.g., the division of the modules is merely a logical function division, and other manners of division may be implemented in practice.
The modules described as separate components may or may not be physically separate, and components shown as modules may or may not be physical units, may be located in one place, or may be distributed over multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional module in the embodiments of the present invention may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units can be realized in a form of hardware or a form of hardware and a form of software functional modules.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential characteristics thereof.
The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference signs in the claims shall not be construed as limiting the claim concerned.
The blockchain is a novel application mode of computer technologies such as distributed data storage, point-to-point transmission, consensus mechanism, encryption algorithm and the like. The blockchain (Blockchain), essentially a de-centralized database, is a string of data blocks that are generated in association using cryptographic methods, each of which contains information from a batch of network transactions for verifying the validity (anti-counterfeit) of its information and generating the next block. The blockchain may include a blockchain underlying platform, a platform product services layer, an application services layer, and the like.
Furthermore, it is evident that the word "comprising" does not exclude other elements or steps, and that the singular does not exclude a plurality. Multiple units or systems as set forth in the system claims may also be implemented by means of one unit or system in software or hardware. The terms second, etc. are used to denote a name, but not any particular order.
Finally, it should be noted that the above-mentioned embodiments are merely for illustrating the technical solution of the present invention and not for limiting the same, and although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications and equivalents may be made to the technical solution of the present invention without departing from the spirit and scope of the technical solution of the present invention.
Claims (10)
1. A method of processing accumulation fund data, comprising:
acquiring macroscopic economic data and public accumulation fund data;
constructing a macroscopic economic index time sequence according to macroscopic economic data and acquiring an initial weight matrix of macroscopic economic indexes, constructing an public accumulation index time sequence according to the public accumulation data and acquiring an initial weight matrix of public accumulation indexes;
Determining an initial weight vector according to the initial weight of the macro economic index and an initial weight matrix of the public accumulation index, constructing a coordination model based on the initial weight vector, the macro economic index time sequence and the public accumulation index time sequence, and obtaining the optimal public accumulation index weight based on the coordination model; determining an accumulation index based on the optimal accumulation index weight;
And inputting the time sequence of the accumulation index as input data to a preset accumulation index cyclic neural grid and carrying out accumulation index prediction.
2. The method of processing the accumulation data as in claim 1 in which,
The indexes of the public accumulation fund data comprise first-order difference of the collection unit quantity, unit collection proportion, unit collection amount, collection base number, amount paid in deposit amount, collection balance, account opening unit quantity, the collection unit quantity completed in the current month, account selling state unit quantity, hold over state unit quantity and normal state unit quantity;
The index of the macro economic data comprises an enterprise scenic index, an enterprise home confidence index, a domestic total production value, a manufacturing purchasing manager index, a non-manufacturing purchasing manager index, an enterprise number and enterprise profit.
3. The method of processing the accumulation data as in claim 1 in which,
The method of obtaining an initial weight matrix for an metric of an accumulation fund comprises,
Classifying indexes of the public accumulation fund data according to the correlation between the public accumulation fund indexes and the enterprise operating conditions and determining the corresponding classification level;
assigning initial weights to indexes of the public accumulation gold data after the classification and the level corresponding to the classification are determined, and taking Dirichlet distribution as prior distribution of public accumulation gold index prediction to obtain an initial weight matrix of the public accumulation gold indexes; wherein, the probability density function of dirichlet distribution is:
Wherein, X is a set of polynary vectors, X= { X 1,X2,X3,……,xk }/>Alpha is a parameter set,/>Α determines the peak and area of the distribution x i, and as α is greater, the distribution of x i on the coordinate axis becomes more and more concentrated in the central area; and the larger alpha i, the larger the corresponding generated x i.
4. The method of processing the accumulation data as in claim 1 in which,
The method further comprises performing stationarity detection on the principal moment index data and the macro-economic index data before acquiring the optimal principal moment index vector based on the coordination model, the method comprising,
Constructing a regression equation between the real index data of the public accumulation fund and the predictive index data of the public accumulation fund by using the OLS;
Carrying out unit root test on the macro economic index time sequence and the accumulation gold index time sequence by utilizing the regression equation;
when it is determined that there is no unit root, a synergistic relationship exists between the macro economic indicator time series and the public accumulation gold indicator time series.
5. The method of processing the accumulation data as in claim 4 in which,
The unit root test of the macro economic index time sequence and the accumulation gold index time sequence is realized by the following formula by utilizing the regression equation:
A unit root exists in H 0：b1 =0, and a synergistic relationship does not exist between the macro economic index time series and the public accumulation gold index time series;
H 1：b1 is not equal to 0, no unit root exists, and a synergistic relationship exists between the macro economic index time sequence and the accumulation gold index time sequence;
Wherein, E t＝b0+b1*∈t-1;∈t is the residual of the regression equation; h 0 is that there is at least one unit root in the time sequence; h 1 is that one unit root does not exist for the hypothetical time series.
6. The method of processing the accumulation data as in claim 1 in which,
Before constructing the time sequence of the accumulation fund index according to the accumulation fund data and acquiring the initial weight matrix of the accumulation fund index, carrying out data dimensionality removal pretreatment on the accumulation fund data by the following formula;
Xscaled＝Xstd*(xmax-xmin)+xmin
Wherein, X is a matrix of the public accumulation gold data according to time sequence, X ij is the data of the ith row and the jth column, X jmin is the minimum value of the data of the jth column of the matrix, and X jmax is the maximum value of the data of the jth column of the matrix. x max is the maximum of the normalized range; x min is the normalized range minimum.
7. The method of claim 1, further comprising preprocessing the macro economic data before constructing a macro economic indicator time series from the macro economic data and acquiring an initial weight matrix of the macro economic indicator:
Filling the macro economic data, and respectively interpolating data among quarter values of the macro economic data by using a linear difference method to obtain macro economic month data;
And respectively interpolating data between two quarters based on the month data to obtain macroscopic economic quarter data.
8. An accumulation fund data processing system, the system comprising:
The data acquisition unit is used for acquiring macroscopic economic data and public accumulation fund data;
The initial weight acquisition unit is used for constructing a macroscopic economic index time sequence according to macroscopic economic data and acquiring an initial weight matrix of the macroscopic economic index, and constructing a public accumulation index time sequence according to the public accumulation data and acquiring an initial weight matrix of the public accumulation index; determining an initial weight vector according to the initial weight of the macro economic index and an initial weight matrix of the public accumulation index, constructing a coordination model based on the initial weight vector, the macro economic index time sequence and the public accumulation index time sequence, and acquiring the optimal public accumulation index weight based on the coordination model; determining an accumulation index based on the optimal accumulation index weight;
And the prediction unit is used for inputting the time sequence of the accumulation index into a preset circulating neural network for accumulation index prediction and performing accumulation index prediction.
9. An electronic device, the electronic device comprising:
at least one processor; and
A memory communicatively coupled to the at least one processor; wherein,
The memory stores instructions executable by the at least one processor to enable the at least one processor to perform the steps in the aggregate data handling method of any of claims 1 to 7.
10. A computer readable storage medium storing a computer program, which when executed by a processor implements the method of processing the public accumulation fund data as claimed in any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311842058.6A CN117993729A (en) | 2023-12-28 | 2023-12-28 | Method and system for processing public accumulation fund data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311842058.6A CN117993729A (en) | 2023-12-28 | 2023-12-28 | Method and system for processing public accumulation fund data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117993729A true CN117993729A (en) | 2024-05-07 |
Family
ID=90888546
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311842058.6A Pending CN117993729A (en) | 2023-12-28 | 2023-12-28 | Method and system for processing public accumulation fund data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117993729A (en) |
-
2023
- 2023-12-28 CN CN202311842058.6A patent/CN117993729A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zekić-Sušac et al. | Predicting energy cost of public buildings by artificial neural networks, CART, and random forest | |
| Ruiz-Torrubiano et al. | A hybrid optimization approach to index tracking | |
| Aytac et al. | Characterization of demand for short life-cycle technology products | |
| JP7140410B2 (en) | Forecasting system, forecasting method and forecasting program | |
| CN111105092B (en) | Data interaction system and method for allocation of medical insurance allowance of hospital | |
| Ayvaz et al. | An integrated LSTM neural networks approach to sustainable balanced scorecard-based early warning system | |
| Chen et al. | MOEA/D with an improved multi-dimensional mapping coding scheme for constrained multi-objective portfolio optimization | |
| CN117235608B (en) | Risk detection method, risk detection device, electronic equipment and storage medium | |
| CN112686470A (en) | Power grid saturation load prediction method and device and terminal equipment | |
| CN113835947B (en) | Method and system for determining abnormality cause based on abnormality recognition result | |
| Yang | Sales Prediction of Walmart Sales Based on OLS, Random Forest, and XGBoost Models | |
| CN111798152A (en) | Intelligent store management method and device | |
| McCluskey et al. | Computer assisted mass appraisal and the property tax | |
| CN117993729A (en) | Method and system for processing public accumulation fund data | |
| CN110110885A (en) | Information forecasting method, device, computer equipment and storage medium | |
| CN115759395A (en) | Training of photovoltaic detection model, detection method of photovoltaic power generation and related device | |
| CN115130924A (en) | Microgrid power equipment asset evaluation method and system under source grid storage background | |
| CN115237970A (en) | Data prediction method, device, equipment, storage medium and program product | |
| CN115204501A (en) | Enterprise evaluation method and device, computer equipment and storage medium | |
| Yaghoubi et al. | Bank Efficiency Forecasting Model Based on the Modern Banking Indicators Using a Hybrid Approach of Dynamic Stochastic DEA and Meta-Heuristic Algorithms | |
| CN113379531A (en) | Bank client deposit total prediction method and device | |
| Abdolbaghi Ataabadi et al. | The effectiveness of the automatic system of fuzzy logic-based technical patterns recognition: Evidence from Tehran stock exchange | |
| CN112132689A (en) | Recommendation method and device based on time sequence factor event | |
| Krüger | Combining density forecasts under various scoring rules: an analysis of UK inflation | |
| Pawełek et al. | The Random Subspace Method in the Prediction of the Risk of Bankruptcy of Companies in Poland1 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |