CN107871174A - A kind of price of medicinal material Forecasting Methodology - Google Patents
A kind of price of medicinal material Forecasting Methodology Download PDFInfo
- Publication number
- CN107871174A CN107871174A CN201610684822.5A CN201610684822A CN107871174A CN 107871174 A CN107871174 A CN 107871174A CN 201610684822 A CN201610684822 A CN 201610684822A CN 107871174 A CN107871174 A CN 107871174A
- Authority
- CN
- China
- Prior art keywords
- price
- medicinal material
- forecasting methodology
- prediction
- methodology according
- 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
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
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0283—Price estimation or determination
Abstract
The present invention relates to internet Chinese medicine field, it discloses a kind of price of medicinal material Forecasting Methodology, comprises the following steps:(A)Price of medicinal material data prepare;(B)Price of medicinal material analysis of Influential Factors;(C)Establish prediction model based on wavelet neural network;(D)Draw price expectation result.The beneficial effects of the invention are as follows:Prediction effect is fine, and precision is higher;Meanwhile transplantability is higher, different cultivars price and its influence factor only need to be analyzed, can test and predict.
Description
Technical field
The present invention relates to internet arena, more particularly to a kind of price of medicinal material Forecasting Methodology.
Background technology
As the gradual maturation of Chinese Medicinal Materials Markets, Chinese Medicinal Materials Markets prospect are boundless.But no matter for Chinese medicine
Plantation or the trade of Chinese medicine, the successful key of Chinese medicine operation are that correctly analysis is carried out to market, it is necessary to market
Demand, cultivated area, yield, market increase driving factors, weather etc. and analyzed, and to market price tendency carry out section
Learn, correctly prediction judges.For Chinese medicine plant peasant household and Chinese medicine trader's business decision provides reference be our companies not
The actual wishes come.But Chinese Medicinal Materials Markets can not all accomplish the prediction to price of medicinal material now.
The content of the invention
In order to solve the problems of the prior art, the invention provides a kind of price of medicinal material Forecasting Methodology, solves existing skill
The problem of price of medicinal material can not being predicted in art.
A kind of price of medicinal material Forecasting Methodology, comprises the following steps:(A)Price of medicinal material data prepare;(B)Price of medicinal material influences
Factor analysis;(C)Establish prediction model based on wavelet neural network;(D)Draw price expectation result.
As a further improvement on the present invention:The step(A)In, collect and analyze medicinal material average selling price over the years,
Collected by years months price and carry out data storage.
As a further improvement on the present invention:The step(B)In, price of medicinal material is judged using grey relevant degree method
With the degree of association of Influence Factors of Price, and then the size of each influence factor for the influence degree of price of medicinal material is analyzed.
As a further improvement on the present invention:The step(B)In, price of medicinal material influence factor is cultivated area, produced per year
Amount, propagandize factor, precipitation, price index number of rural means of production and agricultural product producer price index.
As a further improvement on the present invention:The step(C)In, medicinal material valency is carried out using BP neural network operation method
Lattice are predicted.
As a further improvement on the present invention:The step(C)In, wavelet neural network by input layer, hidden layer and
Output layer forms, and three etale topology structures are arbitrarily to approach Nonlinear Mapping relation;Connection is had no between each neuron of layer.
As a further improvement on the present invention:The step(C)In, establish forecast model and specifically carry out according to the following steps:
Learning sample is inputted, the output of hidden layer and output layer, calculation error and gradient vector are calculated using current network;Study is calculated
Method is judged, is stopped study when the functional value of algorithm setting is less than default accuracy value, is otherwise continued to learn.
As a further improvement on the present invention:The step(D)In, simulation and prediction is carried out by MATLAB instruments, and obtain
Go out medicinal material price expectation result.
The beneficial effects of the invention are as follows:Prediction effect is fine, and precision is higher;Meanwhile transplantability is higher, difference only need to be analyzed
Kind price and its influence factor, it can test and predict.
【Brief description of the drawings】
Fig. 1 is neutral net schematic flow sheet in the present invention;
Fig. 2 is Learning Algorithm schematic diagram of the present invention.
【Embodiment】
With reference to embodiment and accompanying drawing, the present invention is further described.
A kind of price of medicinal material Forecasting Methodology, comprises the following steps:(A)Price of medicinal material data prepare;(B)Price of medicinal material influences
Factor analysis;(C)Establish prediction model based on wavelet neural network;(D)Draw price expectation result.
The step(A)In, collect and analyze medicinal material average selling price over the years, line number of going forward side by side is collected by years months price
According to storage.
The step(B)In, the degree of association of price of medicinal material and Influence Factors of Price is judged using grey relevant degree method,
And then analyze size of each influence factor for the influence degree of price of medicinal material.
The step(B)In, price of medicinal material influence factor is cultivated area, annual production, propagandizes factor, precipitation, agricultural
Means of production price index and agricultural product producer price index.
The step(C)In, price of medicinal material prediction is carried out using BP neural network operation method.
The step(C)In, wavelet neural network is made up of input layer, hidden layer and output layer, three etale topology structures
Arbitrarily to approach Nonlinear Mapping relation;Connection is had no between each neuron of layer.
The step(C)In, establish forecast model and specifically carry out according to the following steps:Learning sample is inputted, utilizes current net
Network calculates the output of hidden layer and output layer, calculation error and gradient vector;Learning algorithm is judged, when algorithm setting
Functional value stops study when being less than default accuracy value, otherwise continues to learn.
The step(D)In, simulation and prediction is carried out by MATLAB instruments, and draw price of medicinal material prediction result.
By collecting and analyzing hawthorn average selling price and influence factor over the years, using data mining to receive carry out data
Prepare, establish data mining model, data prediction and conclusion statement, reach this with research so as to carrying out a series of study and grind
The expection studied carefully.Initial data is pre-processed first, and each influence factor and valency are analyzed using grey relational grade analysis method
The degree of association of lattice, so that it is determined that key influence factor.The method of wavelet neural network again, by MATLAB instruments emulate pre-
Survey, draw prediction result.
In one embodiment, we are illustrated using the price of hawthorn.
During Data Collection, the price of hawthorn we have selected the hawthorn moon price of 2004 to 2014, its influence factor
For annual data, it is identical dimension to keep data, is annual average evidence by the monthly average data processing of hawthorn.On time
Between, year average price/member, cultivated area/ten thousand mu, Disaster Area/hectare, annual production, propagation factor and temperature on average enter line number
According to collection;Meanwhile also temporally, sunshine time, mean wind speed, average relative humidity, precipitation, prices of the means of agricultural production
Index and agricultural product producer price index carry out hawthorn Data Collection.
Influence the correlation analysis of price factor:Because the Influence Factors of Price of selection is more, in order to judge these factors
Whether it is key influence factor, if redundancy be present, we judge hawthorn price and each factor with grey relevant degree method
The degree of association, and then the size of each influence factor for the influence degree of hawthorn price is analyzed, according to hawthorn price and each influence
The degree of association index of factor, when degree of association index more than 0.6 factor for example initial selected cultivated area, annual production, propagandize factor,
6 precipitation, price index number of rural means of production, agricultural product producer price index indexs refer to as our initial modeling
Mark.Complexity and index in view of model is to the influence degree of price, and we are in price index number of rural means of production and agricultural production
One index of selection selects degree of association index larger as our modeling index, herein we in product producer price index
The modeling index as us, therefore the index of final modeling is:Cultivated area, annual production, propagandize factor, precipitation, agriculture
5 indexs of production person price index.
Prediction model based on wavelet neural network:
After to influenceing price factor analysis, Chinese Medicinal Materials Markets are found as the changeable environment of dynamic, there is substantial amounts of non-thread
Property, uneven stability, uncertainty and time variation.Because traditional time series models are non-for processing dynamic non-stationary, dynamic
The limitation of linear problem, centering price of medicinal material prediction effect are not ideal.However, in the computational methods of intelligence, artificial god
It is with a kind of method for representing meaning through network, has single neuron to form network, although it is relatively simple, but can embody non-
The function of linear nature's feature, due to this ability so that mould is established with non-linear continuous function by artificial neural network
Type is possibly realized.In artificial neural network, BP neural network operation method simple utilization is the most extensive, but pre- in hawthorn price
Survey model in error compared with.Wavelet neural network is combined neutral net with the advantages of Wavelet Analysis Theory, and in research trends
During the problem of non-stationary, kinematic nonlinearity, the contraction-expansion factor and shift factor newly introduced, make wavelet neural network in practical application
In approximation capability more strengthen, precision of prediction is high, strong applicability.So it is secondary we to select wavelet neural network to establish hawthorn pre-
Survey model.
Neutral net is made up of Three Tiered Network Architecture, as input layer, hidden layer, output layer, between each layer uniformly
Node.The feature of the system of neutral net allows three etale topology structures arbitrarily to approach Nonlinear Mapping relation.Structure below figure.
The neuron of each layer is widely coupled with next layer of all neurons, but has no connection between each neuron of layer,
The direction of data processing be arrow represent direction, such as Fig. 2.
N=5 in the modeling process of hawthorn, i.e., above-named 5 indexs, the year price of m=1, as hawthorn.
The specific implementation step of learning algorithm is played as follows:(1)The initialization of network parameter.(2)It is corresponding to input learning sample
Desired output.(3)The output of hidden layer and output layer is calculated using current network.(4)Calculation error and gradient vector.(5)
Change network parameter values.(6)Whether evaluation algorithm terminates.
Work as E<When, i.e. function E is less than some accuracy value set in advance, stop the study of network, otherwise step(2)
Circulation.
The model is tested, in one embodiment, the data of 2004 to 2012 is have selected and carries out network training
(As training set), the data of 2013 and 2014 are as the test to network(As test set)Tested, its result
In 9 groups of training samples prediction curve and actual value price curve trend it is basically identical, and the desired value of 2 groups of test samples
It is smaller with the gap of actual value, reach expected test effect.
Forecast model:
In one embodiment, selection is so obtain data(2004 to 2014)As training sample, the network trained is entered
The row price expectation of 2015.
Its output result is better with the Approximation effect for increasing network of training sample, 11 groups of training samples it is pre-
The price curve trend of survey curve and actual value is basically identical, and effect is preferable, and forecast price and real price are much like.
Wavelet neural network is fine as hawthorn price expectation effect, and precision is higher.Transplantability is higher, only need to analyze difference
Kind price and its influence factor, it can bring the model into and test and predict.
Above content is to combine specific preferred embodiment further description made for the present invention, it is impossible to is assert
The specific implementation of the present invention is confined to these explanations.For general technical staff of the technical field of the invention,
On the premise of not departing from present inventive concept, some simple deduction or replace can also be made, should all be considered as belonging to the present invention's
Protection domain.
Claims (8)
1. a kind of price of medicinal material Forecasting Methodology, it is characterised in that comprise the following steps:(A)Price of medicinal material data prepare;(B)Medicine
Material Influence Factors of Price is analyzed;(C)Establish prediction model based on wavelet neural network;(D)Draw price expectation result.
2. price of medicinal material Forecasting Methodology according to claim 1, it is characterised in that:The step(A)In, collect and analyze
Medicinal material average selling price over the years, is collected by years months price and carries out data storage.
3. price of medicinal material Forecasting Methodology according to claim 2, it is characterised in that:The step(B)In, closed using grey
Connection degree method judges the degree of association of price of medicinal material and Influence Factors of Price, and then analyzes each influence factor for price of medicinal material
The size of influence degree.
4. price of medicinal material Forecasting Methodology according to claim 1, it is characterised in that:The step(B)In, price of medicinal material shadow
The factor of sound is cultivated area, annual production, propagandizes factor, precipitation, price index number of rural means of production and agricultural product producer's valency
Grid index.
5. price of medicinal material Forecasting Methodology according to claim 1, it is characterised in that:The step(C)In, using BP nerves
Network operations method carries out price of medicinal material prediction.
6. price of medicinal material Forecasting Methodology according to claim 1, it is characterised in that:The step(C)In, Wavelet Neural Network
Network is made up of input layer, hidden layer and output layer, and three etale topology structures are arbitrarily to approach Nonlinear Mapping relation;With each god of layer
Through having no connection between member.
7. price of medicinal material Forecasting Methodology according to claim 1, it is characterised in that:The step(C)In, establish prediction mould
Type is specifically carried out according to the following steps:Learning sample is inputted, the output of hidden layer and output layer is calculated using current network, calculates and misses
Difference and gradient vector;Learning algorithm is judged, stops study when the functional value of algorithm setting is less than default accuracy value,
Otherwise continue to learn.
8. price of medicinal material Forecasting Methodology according to claim 1, it is characterised in that:The step(D)In, pass through MATLAB
Instrument carries out simulation and prediction, and draws price of medicinal material prediction result.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610684822.5A CN107871174A (en) | 2016-09-23 | 2016-09-23 | A kind of price of medicinal material Forecasting Methodology |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610684822.5A CN107871174A (en) | 2016-09-23 | 2016-09-23 | A kind of price of medicinal material Forecasting Methodology |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107871174A true CN107871174A (en) | 2018-04-03 |
Family
ID=61750567
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610684822.5A Pending CN107871174A (en) | 2016-09-23 | 2016-09-23 | A kind of price of medicinal material Forecasting Methodology |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107871174A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110298681A (en) * | 2019-04-24 | 2019-10-01 | 内蒙古科技大学 | A kind of price expectation method |
CN114444946A (en) * | 2022-01-28 | 2022-05-06 | 荃豆数字科技有限公司 | Traditional Chinese medicine decoction piece refined operation guidance method and device and computer equipment |
-
2016
- 2016-09-23 CN CN201610684822.5A patent/CN107871174A/en active Pending
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110298681A (en) * | 2019-04-24 | 2019-10-01 | 内蒙古科技大学 | A kind of price expectation method |
CN114444946A (en) * | 2022-01-28 | 2022-05-06 | 荃豆数字科技有限公司 | Traditional Chinese medicine decoction piece refined operation guidance method and device and computer equipment |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Liu et al. | A future land use simulation model (FLUS) for simulating multiple land use scenarios by coupling human and natural effects | |
Xu et al. | Design of an integrated climatic assessment indicator (ICAI) for wheat production: A case study in Jiangsu Province, China | |
Özger et al. | Comparison of wavelet and empirical mode decomposition hybrid models in drought prediction | |
Huo et al. | Artificial neural network models for reference evapotranspiration in an arid area of northwest China | |
CN101480143B (en) | Method for predicating single yield of crops in irrigated area | |
Blanc | Statistical emulators of maize, rice, soybean and wheat yields from global gridded crop models | |
Kankal et al. | Modeling and forecasting of Turkey’s energy consumption using socio-economic and demographic variables | |
Safa et al. | Determination and modelling of energy consumption in wheat production using neural networks:“A case study in Canterbury province, New Zealand” | |
CN107230113A (en) | A kind of house property appraisal procedure of multi-model fusion | |
Dou et al. | Estimating forest carbon fluxes using four different data-driven techniques based on long-term eddy covariance measurements: Model comparison and evaluation | |
Lu et al. | Evaluation the effect of cultivated land protection policies based on the cloud model: A case study of Xingning, China | |
CN102541920A (en) | Method and device for improving accuracy degree by collaborative filtering jointly based on user and item | |
CN106022954A (en) | Multiple BP neural network load prediction method based on grey correlation degree | |
Fang et al. | Application of long short-term memory (LSTM) on the prediction of rainfall-runoff in karst area | |
CN105184400A (en) | Tobacco field soil moisture prediction method | |
Hjelkrem et al. | A process-based model to forecast risk of potato late blight in Norway (The Nærstad model): model development, sensitivity analysis and Bayesian calibration | |
Rocha et al. | Early prediction of durum wheat yield in Spain using radial basis functions interpolation models based on agroclimatic data | |
CN107871174A (en) | A kind of price of medicinal material Forecasting Methodology | |
Liu et al. | Evaluation and analysis of irrigation water use efficiency based on an extreme learning machine model optimized by the spider monkey optimization algorithm | |
CN109190810A (en) | The prediction technique of grassland in northern China area NDVI based on TDNN | |
Wu et al. | Calculation and analysis of agricultural carbon emission efficiency considering water–energy–food pressure: Modeling and application | |
Liu et al. | Simulations of runoff and evapotranspiration in Chinese fir plantation ecosystems using artificial neural networks | |
CN109272144A (en) | The prediction technique of grassland in northern China area NDVI based on BPNN | |
CN105046089A (en) | Method for predicting strong rainfall and flood disasters | |
CN107578130A (en) | A kind of method of the copper forecasting of futures prix of the TSK fuzzy systems of structure and parameter combination learning |
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 | ||
WD01 | Invention patent application deemed withdrawn after publication | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20180403 |