CN110009405A - A kind of passenger car sales volume simulating and predicting method based on network generalized extreme value model - Google Patents
A kind of passenger car sales volume simulating and predicting method based on network generalized extreme value model Download PDFInfo
- Publication number
- CN110009405A CN110009405A CN201910222064.9A CN201910222064A CN110009405A CN 110009405 A CN110009405 A CN 110009405A CN 201910222064 A CN201910222064 A CN 201910222064A CN 110009405 A CN110009405 A CN 110009405A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- sales volume
- model
- extreme value
- network
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/231—Hierarchical techniques, i.e. dividing or merging pattern sets so as to obtain a dendrogram
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0202—Market predictions or forecasting for commercial activities
Abstract
Passenger car sales volume simulating and predicting method provided in an embodiment of the present invention based on network generalized extreme value model, network generalized extreme value model is input to by the parameter information for the vehicle to be predicted for inputting user, the sales volume of the prediction vehicle to be predicted that can be more accurate, and use network generalized extreme value model, the quantity for needing to be estimated coefficient can be greatly reduced, not only reduce the requirement to data sample size, also effectively improves model accuracy.
Description
Technical field
The present invention relates to passenger car sales volume analysis prediction field more particularly to a kind of multiplying based on network generalized extreme value model
With vehicle sales volume simulating and predicting method.
Background technique
Passenger car Method for Sales Forecast technology refers to establishing model according to previous sales data and other data, and under
The sales volume in some a stage is estimated.Existing passenger car Method for Sales Forecast technology mainly uses time series models (Time-
Series Model): the rule that sales volume changes is decomposed into trend part, seasonal part, autoregression with time series models
Part；When prediction, following sales volume of prediction is deduced according to the rule of 3 parts；With log-linear model (Log-linear
Model): the driving factors data and quantization causality of sales volume variation are found with log-linear regression model；When prediction, first
It predicts the value of all driving factors data, sales volume is then predicted according to the causality of quantization.
However, given a forecast using time series models from historical law, sales volume wave caused by no method interpretation factor
It is dynamic, the especially driving factors data that did not occurred of history, such as event of burst, the policy newly put into effect；Using logarithmic linear mould
It is explanatory to solve the problems, such as that time series models lack for type, but since the factor for influencing sales volume is very more, so that model is
Number is excessive and is difficult to be estimated.
Summary of the invention
The purpose of the embodiment of the present invention is that providing a kind of passenger car sales volume simulation and forecast based on network generalized extreme value model
Method can solve the problems, such as that the quantity of model coefficient is excessive, substantially increase pin under the premise of service factor driving model
Measure the efficiency of prediction.
To achieve the above object, the embodiment of the invention provides a kind of passenger car sales volumes based on network generalized extreme value model
Simulating and predicting method, comprising the following steps:
Obtain the parameter information of the vehicle to be predicted of user's input；
According to the parameter information, the first driving factors data of the vehicle to be predicted are extracted；
The pin of the vehicle to be predicted is obtained using network generalized extreme value model according to the first driving factors data
The amount of selling.
Further, the network generalized extreme value model constructs by the following method:
Obtain the parameter information and sales volume of more candidates vehicle similar with the vehicle attribute to be predicted；
By calculating the alternative between described more candidate vehicles, the network structure of candidate vehicle is obtained；
According to the parameter information of described more candidate vehicles, the second driving factors data of described more candidate vehicles are extracted；
According to the second driving factors data and the candidate network structure of vehicle and the pin of the more candidate vehicles
Amount, obtains the model coefficient of preset model using gradient descent algorithm；
According to the model structure of the preset model and the model coefficient, the network generalized extreme value model is constructed.
Further, the preset model structure is fi=αi+β1i* pricei+β2i* Brandi+β3i* product capabilityi+
β4i* vitalityi+β5iChannel poweri+β6i* marketing influencei；
Wherein, fiFor vehicleiSales volume, pricei, Brandi, product capabilityi, vitalityi, channel poweriAnd marketing influences
PoweriFor vehicleiDriving factors data, αi、β1i、β2i、β3i、β4i、β5iAnd β6iFor the model coefficient.
Further, the alternative by calculating between described more candidate vehicles, obtains the network structure of candidate vehicle,
Specifically:
The alternative between described more candidate vehicles is calculated according to the following formula:
Wherein, vehiclekIt is the vehicle of candidate vehicle for the vehicle of vehicle to be measured, the vehicle i and vehicle j；
Nested structure is obtained using hierarchy clustering method, and the nested structure is adjusted according to qualitative and Quantitative algorithm
It is whole, obtain the network structure of the candidate vehicle.
Further, the driving factors data include price, Brand, product capability, vitality, channel power and marketing
Influence power；
The calculation method of the price are as follows:
The calculation method of the price are as follows: use model sales volume to weight as weight to model knock-down price, and do mobile flat
It handles, obtains the price.
Further, the calculation method of the Brand are as follows:
It calculates the brand and accounts for the business share of the whole city, and do large span rolling average processing, obtain the Brand.
Further, the calculation method of the product capability are as follows:
Using product attribute and consumers' perceptions data, composite calulation is carried out to multiple sub-indicators of product, obtains institute
State product capability.
Further, the calculation method of the vitality are as follows:
Vehicle is calculated in the vitality of different times by class index curve；Wherein, the different times are divided into vehicle
Vitality of the age in 6 months climbs the phase and vehicle age is more than 6 months vitality decline phases.
Further, the calculation method of the channel power are as follows:
The Intrusion Index that city site is weighted by sales volume share, obtains the channel power.
Further, the calculation method of the marketing influence are as follows:
It is calculated by the rolling average of network attention index, obtains the marketing influence.
Compared with prior art, it has the following beneficial effects:
Passenger car sales volume simulating and predicting method provided in an embodiment of the present invention based on network generalized extreme value model, pass through by
The parameter information of the vehicle to be predicted of user's input is input to network generalized extreme value model, described in prediction that can be more accurate to
The sales volume of prediction vehicle can be greatly reduced the quantity for needing to be estimated coefficient, not only be subtracted using network generalized extreme value model
Lack the requirement to data sample size, also effectively improves model accuracy.
Detailed description of the invention
Fig. 1 is a reality of the passenger car sales volume simulating and predicting method provided by the invention based on network generalized extreme value model
Apply the flow diagram of example；
Fig. 2 is a kind of flow diagram of embodiment of network generalized extreme value model building method provided by the invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
It is the passenger car sales volume simulating and predicting method provided by the invention based on network generalized extreme value model referring to Fig. 1, Fig. 1
One embodiment flow diagram；The embodiment of the present invention provides a kind of passenger car sales volume based on network generalized extreme value model
Simulating and predicting method, including step S1 to S3；
S1 obtains the parameter information of the vehicle to be predicted of user's input.
S2 extracts the first driving factors data of the vehicle to be predicted according to the parameter information.
It should be noted that driving factors data include price, Brand, product capability, vitality, channel power and marketing
Influence power.
Wherein, price is to represent the monetary cost that consumer pays with vehicle knock-down price；Brand is that consumer is enjoyed
The emotional satisfaction arrived is presented as " more people's purchases ", " class is high " in Chinese market；Product capability is to represent consumer's harvest
Material enjoyment, mainly embodied with the perception of consumer, meanwhile, and be able to reflect out product attribute；Vitality is to represent vehicle
The newness degree of type, the short vehicle of the usual preference Time To Market of consumer, appearance have more feeling of freshness, and technical equipment is more novel, more
With technology sense；Channel power purchases a possibility that vehicle provides purchase and convenience for consumer；Marketing influence is marketing activity pair
Consumer buys to be influenced caused by the wish of vehicle.
S3 obtains the vehicle to be predicted using network generalized extreme value model according to the first driving factors data
Sales volume.
Under the premise of service factor driving model, using network generalized extreme value model, it is able to solve the number of model coefficient
Measure excessive problem.Such as assume there is 3 vehicles A, B, C, monthly sales volume is logarithmic linear by 2 factor prices, brand influences
The model structure of model is as follows:
Sales volumeA=αA+β1A* priceA+β2A* brandA+β3A* priceB+β4A* brandB+β5A* priceC+β6A* brandC；
Sales volumeB=αB+β1B* priceA+β2B* brandA+β3B* priceB+β4B* brandB+β5B* priceC+β6B* brandC；
Sales volumeC=αC+β1C* priceA+β2C* brandA+β3C* priceB+β4C* brandB+β5C* priceC+β6C* brandC；
Above formula shares 3*3*2+3=21 coefficient.If establishing log-linear model to 20 vehicles, 10 factors,
There are coefficient 20*20*10+20=4020, such quantity makes coefficient estimation become extremely difficult.
And network generalized extreme value model (Network Generalized Extreme Value Model) uses network knot
Structure expresses the competition between vehicle.If indicating that this network structure, above-mentioned model structure simplify with f are as follows:
fA(sales volumeA, sales volumeB, sales volumeC)=αA+β1A* priceA+β2A* brandA；
fB(sales volumeA, sales volumeB, sales volumeC)=αB+β1B* priceB+β2B* brandB；
fC(sales volumeA, sales volumeB, sales volumeC)=αC+β1C* priceC+β2C* brandC；
Above formula shares 3*2+3=9 coefficient.For the scene of 20 vehicles, 20*10+20=is greatly reduced in coefficient
220.
So the present invention uses network generalized extreme value model, enable to the prediction to passenger car sales volume more accurate,
Improve the efficiency of passenger car Method for Sales Forecast.
Fig. 2 is referred to, Fig. 2 is a kind of process of embodiment of network generalized extreme value model building method provided by the invention
Schematic diagram；Network generalized extreme value model building method, including step S10-S14；
S10 obtains the parameter information and sales volume of more candidates vehicle similar with the vehicle attribute to be predicted.
S11 obtains the network structure of candidate vehicle by calculating the alternative between described more candidate vehicles.
In the present embodiment, the network structure of candidate vehicle is obtained by following steps:
It calculates alternative between vehicle:
Wherein, vehiclekIt is the vehicle of candidate vehicle for the vehicle of vehicle to be measured, the vehicle i and vehicle j；
Nested structure is obtained using hierarchy clustering method, and using nested structure as initial scheme, and according to qualitative and fixed
Quantity algorithm is adjusted the nested structure, obtains the network structure of the candidate vehicle.
S12 extracts the second driving factors number of described more candidate vehicles according to the parameter information of described more candidate vehicles
According to.
In embodiments of the present invention, the model structure of preset model is fi=αi+β1i* pricei+β2i* Brandi+β3i*
Product capabilityi+β4i* vitalityi+β5iChannel poweri+β6i* marketing influencei；
Wherein, fiFor vehicleiSales volume, pricei, Brandi, product capabilityi, vitalityi, channel poweriAnd marketing influences
PoweriFor vehicleiDriving factors data, αi、β1i、β2i、β3i、β4i、β5iAnd β6iFor the model coefficient.
S13, according to the network structure and the more candidate vehicles of the second driving factors data and the candidate vehicle
Sales volume, obtain the model coefficient of preset model using gradient descent algorithm.
Wherein, gradient descent algorithm is to solve for the commonly used method of machine learning model parameter.Solving loss function
Minimum value when, can by gradient descent algorithm come progressive alternate solution, the loss function and model parameter minimized
Value.
In the present embodiment, the price of driving factors data obtains in the following manner: using model sales volume as weight
Model knock-down price is weighted, and does rolling average processing, obtains the price.
As the preferred embodiment of the present invention specifically: according to the sales volume of the N the end of month month knock-down price and the N month of vehicle, obtain N
Month vehicle the end of month knock-down price, whereinAccording to institute
The N vehicle month, the knock-down price the end of month is stated, N month vehicle knock-down price is obtained, wherein Using the N month vehicle knock-down price as the price.
The Brand of driving factors data obtains in the following manner: the business share that the brand accounts for the whole city is calculated, and
Large span rolling average processing is done, the Brand is obtained.
As the preferred embodiment of the present invention specifically: according to the sales volume and the guiding price of the N month of the N month of vehicle, obtain the N month
Brand market value share, wherein describedAccording to
Past 24 months brand market value shares obtain N month Brand=past 24 months brand market value share average value.
The product capability of driving factors data obtains in the following manner: product attribute and consumers' perceptions data are used, it is right
Multiple sub-indicators of product carry out composite calulation, obtain the product capability.
As the preferred embodiment of the present invention specifically: according to vehicle in the scoring of the family of automobile and the perception of consumer,
Obtain N month product capability, wherein the N month product capability=p1* space+p2* power+p3* manipulation+p4* oil consumption+p5* comfort+p6*
Interior trim+p7* safety+p8* exterior arrangement+p9* appearance.
It should be noted that the p1-p9It is to be obtained by consumer evaluation's data on statistics investigational data or line.
Wherein, data cases are from the configuration information of vehicle, the scoring of the family of investigation CPV value and automobile, and pass through
Physical parameter conversion and configuration CPV aggregation, are calculated each data, as shown in the table:
The vitality of driving factors data obtains in the following manner: vehicle is calculated in difference by class index curve
The vitality in period；Wherein, the different times are divided into that vitality of the vehicle age in 6 months climbs the phase and vehicle age is more than 6
The vitality decline phase of the moon.
As the preferred embodiment of the present invention specifically: when vehicle Time To Market is in 6 months, N month vitality=[1-
exp(-c1* (vehicle age+1)+c2)]*c3+c4；
When vehicle Time To Market is more than 6 months, N month vitality=exp [(- d1* (vehicle age -3)+d2^5*d3+d4.
The channel power of driving factors data obtains in the following manner: being referred to by the influence that sales volume share weights city site
Number, obtains the channel power.
As the preferred embodiment of the present invention specifically: sales quota and the city site according to vehicle in city
Intrusion Index obtains N month channel power, wherein N month channel power=∑ (the monthly city sales volume share × month city N site shadow
Snap number.
The marketing influence of driving factors data obtains in the following manner: passing through the rolling average meter of network attention index
It calculates, obtains the marketing influence.
As the preferred embodiment of the present invention specifically: according to the concern index of vehicle, N month marketing influence is obtained,
In, it is described
According to the above-mentioned price being calculated, Brand, product capability, vitality, channel power, marketing influence 6 drivings
Factor data and sales volume fi, model coefficient is obtained using gradient descent algorithm.
S14 constructs the network generalized extreme value mould according to the model structure of the preset model and the model coefficient
Type.
Finally according to the parameter information of vehicle to be predicted, the driving factors of the vehicle to be predicted are extracted, and are sent to
The network generalized extreme value model built, obtains the Method for Sales Forecast result of the vehicle to be predicted.
As a preferred embodiment of the present invention, embodiment provided by the invention has " Gao Fang in terms of application value
Very, multi-direction, multi-level " advantage:
1, high emulation
Embodiment provided by the invention is by (price, product capability, vitality, channel power, is marketed at Brand in automobile market
Influence power, competitive relation), outer (macroeconomy, Seasonal Regularity) multiple important factor in order quantizations, simulating some automobile product
There is true to nature, dynamic advantage when operation law in market:
(1), reduced time series model (Time-series Model) can provide the impact analysis for influencing sales volume factor；
(2), comparison log-linear model (Log-linear Model) can be under the premise of reducing coefficient estimation difficulty, more
The competitive relation neatly provided between each money vehicle of automobile market influences.
2, multi-direction
After each factors quantization, embodiment provided by the invention can be applied from positive, reverse both direction:
(1), positive: can be changed by presetting some vehicle in automobile market intrinsic factor, carry out sales volume simulation.User
It can be converted into market information after obtaining sales volume analog result and be used to the links such as produce, invest.
(2), reverse: can be changed by the sales volume and market segment intrinsic factor for presetting some vehicle, carry out remaining market
The simulation of intrinsic factor.Such as, user can be in pre-designed cross out amount and price, Brand, product capability, channel power, marketing shadow
After the variation for ringing power, whether simulation needs further to be managed the life cycle of vehicle.
3, multi-level
Embodiment provided by the invention can in conjunction with a variety of usage modes, be different user group'ss " information ", " method ",
" tool " three levels provide service:
(1), scheme is applied to specific certain vehicles, and according to preset each influence factor, is provided for user
Direct Market Simulation information；
(2), the default factor for being best suitable for market situation in (1) is selected as the case where being most likely to occur, and scheme is acted on
To the Method for Sales Forecast of specific certain vehicles, direct market prediction information is provided for user；
(3), scheme is applied to specific certain vehicles, user according to the relationship of influence factor and sales volume, by from
The default influence factor of row, provides the analog information under a variety of scenes for user；
(4), by method combining information technology in (3), real-time mould can be provided for user based on mobile phone, PC machine platform
Quasi- forecasting system:
I), using cloud server as operation terminal, using the end PC web interface as the cloud computing system of client；
Ii), price, Brand, product capability, vitality, channel power, marketing influence this six factors are provided for client
Controllable components allow the detailed-oriented control variation ratio of user, so as to adjust every predictive factors；
Iii), according to the default factor of items, carry out cloud and calculate in real time, simulated in real time, prediction result.
In conclusion the passenger car sales volume simulation and forecast side provided in an embodiment of the present invention based on network generalized extreme value model
Method, the parameter information of the vehicle to be predicted by obtaining user's input, the driving factors data that will be extracted from parameter information
It is input to network generalized extreme value model, can accurately predict the sales volume of vehicle to be predicted.In addition, using network generalized extreme value mould
Type after the quantity for needing to be estimated coefficient capable of being greatly reduced, not only reduces the requirement to data sample size, also effectively
Improve model accuracy.
Those of ordinary skill in the art will appreciate that realizing all or part of the process in above-described embodiment method, being can be with
Relevant hardware/relevant device/related system is instructed to complete by computer program, the equipment includes: processor, deposits
Reservoir and storage are in the memory and the computer program that can run on the processor.The processor executes institute
Above-mentioned each passenger car sales volume simulating and predicting method embodiment based on network generalized extreme value model is realized when stating computer program
In step, such as step S1 to S3 shown in FIG. 1.
Alleged processor can be central processing unit (Central Processing Unit, CPU), can also be it
His general processor, digital signal processor (Digital Signal Processor, DSP), specific integrated circuit
(Application Specific Integrated Circuit, ASIC), ready-made programmable gate array (Field-
Programmable Gate Array, FPGA) either other programmable logic device, discrete gate or transistor logic,
Discrete hardware components etc..General processor can be microprocessor or the processor is also possible to any conventional processor
Deng, the processor is the control centre of the passenger car sales volume simulation and forecast equipment based on network generalized extreme value model, benefit
Entire each portion of the passenger car sales volume simulation and forecast equipment based on network generalized extreme value model with various interfaces and connection
Point.
The memory can be used for storing the computer program and/or module, and the processor is by operation or executes
Computer program in the memory and/or module are stored, and calls the data being stored in memory, described in realization
The various functions of Method for Sales Forecast equipment/system of passenger car.The memory can mainly include storing program area and storing data
Area, wherein storing program area can application program needed for storage program area, at least one function (such as sound-playing function,
Image player function etc.) etc.；Storage data area, which can be stored, uses created data (such as audio data, electricity according to mobile phone
Script for story-telling etc.) etc..In addition, memory may include high-speed random access memory, it can also include nonvolatile memory, such as
Hard disk, memory, plug-in type hard disk, intelligent memory card (Smart Media Card, SMC), secure digital (Secure
Digital, SD) card, flash card (Flash Card), at least one disk memory, flush memory device or other volatibility are solid
State memory device.
Wherein, if module/unit of the sales volume simulation and forecast integration of equipments of the passenger car is with SFU software functional unit
Form realize and when sold or used as an independent product, can store in a computer readable storage medium.Base
In such understanding, the present invention realizes all or part of the process in above-described embodiment method, can also pass through computer program
It is completed to instruct relevant hardware, the computer program can be stored in a computer readable storage medium, the calculating
Machine program is when being executed by processor, it can be achieved that the step of above-mentioned each embodiment of the method.Wherein, the computer program includes
Computer program code, the computer program code can for source code form, object identification code form, executable file or certain
A little intermediate forms etc..The computer-readable medium may include: any entity that can carry the computer program code
Or device, recording medium, USB flash disk, mobile hard disk, magnetic disk, CD, computer storage, read-only memory (ROM, Read-Only
Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software
Distribution medium etc..
It should be noted that equipment/system embodiment described above is only schematical, wherein described be used as is divided
Unit from part description may or may not be physically separated, component shown as a unit can be or
It may not be physical unit, it can it is in one place, or may be distributed over multiple network units.It can basis
It is actual to need that some or all of the modules therein is selected to achieve the purpose of the solution of this embodiment.Ordinary skill people
Member can understand and implement without creative efforts.
The above is a preferred embodiment of the present invention, it is noted that for those skilled in the art
For, various improvements and modifications may be made without departing from the principle of the present invention, these improvements and modifications are also considered as
Protection scope of the present invention.
Claims (10)
1. a kind of passenger car sales volume simulating and predicting method based on network generalized extreme value model, which is characterized in that including following step
It is rapid:
Obtain the parameter information of the vehicle to be predicted of user's input；
According to the parameter information, the first driving factors data of the vehicle to be predicted are extracted；
The sales volume of the vehicle to be predicted is obtained using network generalized extreme value model according to the first driving factors data.
2. as described in claim 1 based on the passenger car sales volume simulating and predicting method of network generalized extreme value model, feature exists
In the network generalized extreme value model constructs by the following method:
Obtain the parameter information and sales volume of more candidates vehicle similar with the vehicle attribute to be predicted；
By calculating the alternative between described more candidate vehicles, the network structure of candidate vehicle is obtained；
According to the parameter information of described more candidate vehicles, the second driving factors data of described more candidate vehicles are extracted；
According to the second driving factors data and the candidate network structure of vehicle and the sales volume of the more candidate vehicles,
The model coefficient of preset model is obtained using gradient descent algorithm；
According to the model structure of the preset model and the model coefficient, the network generalized extreme value model is constructed.
3. as claimed in claim 2 based on the passenger car sales volume simulating and predicting method of network generalized extreme value model, feature exists
In the model structure of the preset model is fi=αi+β1i* pricei+β2i* Brandi+β3i* product capabilityi+β4i* vitalityi+
β5iChannel poweri+β6i* marketing influencei；
Wherein, fiFor vehicleiSales volume, pricei, Brandi, product capabilityi, vitalityi, channel poweriAnd marketing influenceiFor
VehicleiDriving factors data, αi、β1i、β2i、β3i、β4i、β5iAnd β6iFor the model coefficient.
4. as claimed in claim 2 based on the passenger car sales volume simulating and predicting method of network generalized extreme value model, feature exists
In, the alternative by calculating between described more candidate vehicles obtains the network structure of candidate vehicle, specifically:
The alternative between described more candidate vehicles is calculated according to the following formula:
Wherein, vehiclekIt is the vehicle of candidate vehicle for the vehicle of vehicle to be measured, the vehicle i and vehicle j；
Nested structure is obtained using hierarchy clustering method, and the nested structure is adjusted according to qualitative and Quantitative algorithm,
Obtain the network structure of the candidate vehicle.
5. the passenger car sales volume simulating and predicting method as described in any one of claims 1-3 based on network generalized extreme value model,
It is characterized in that, the driving factors data include that price, Brand, product capability, vitality, channel power and marketing influence
Power；
The calculation method of the price are as follows: use model sales volume to weight as weight to model knock-down price, and do at rolling average
Reason, obtains the price.
6. as claimed in claim 5 based on the passenger car sales volume simulating and predicting method of network generalized extreme value model, feature exists
In the calculation method of the Brand are as follows:
It calculates the brand and accounts for the business share of the whole city, and do large span rolling average processing, obtain the Brand.
7. as claimed in claim 5 based on the passenger car sales volume simulating and predicting method of network generalized extreme value model, feature exists
In the calculation method of the product capability are as follows:
Using product attribute and consumers' perceptions data, composite calulation is carried out to multiple sub-indicators of product, obtains the production
Product power.
8. as claimed in claim 5 based on the passenger car sales volume simulating and predicting method of network generalized extreme value model, feature exists
In the calculation method of the vitality are as follows:
Vehicle is calculated in the vitality of different times by class index curve；Wherein, the different times are divided into vehicle age 6
Vitality in a month climbs the phase and vehicle age is more than 6 months vitality decline phases.
9. as claimed in claim 5 based on the passenger car sales volume simulating and predicting method of network generalized extreme value model, feature exists
In the calculation method of the channel power are as follows:
The Intrusion Index that city site is weighted by sales volume share, obtains the channel power.
10. as claimed in claim 5 based on the passenger car sales volume simulating and predicting method of network generalized extreme value model, feature exists
In the calculation method of the marketing influence are as follows:
It is calculated by the rolling average of network attention index, obtains the marketing influence.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910222064.9A CN110009405A (en) | 2019-03-22 | 2019-03-22 | A kind of passenger car sales volume simulating and predicting method based on network generalized extreme value model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910222064.9A CN110009405A (en) | 2019-03-22 | 2019-03-22 | A kind of passenger car sales volume simulating and predicting method based on network generalized extreme value model |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110009405A true CN110009405A (en) | 2019-07-12 |
Family
ID=67167788
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910222064.9A Pending CN110009405A (en) | 2019-03-22 | 2019-03-22 | A kind of passenger car sales volume simulating and predicting method based on network generalized extreme value model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110009405A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110555730A (en) * | 2019-08-28 | 2019-12-10 | 上海明品医学数据科技有限公司 | Data statistical analysis method for product after marketing research |
| CN111401941A (en) * | 2020-03-06 | 2020-07-10 | 武汉大学 | Vehicle sales prediction method based on XGboost recommendation algorithm |
| CN113554183A (en) * | 2021-08-03 | 2021-10-26 | 同济大学 | Extreme value prediction method based on unsupervised machine learning algorithm |
-
2019
- 2019-03-22 CN CN201910222064.9A patent/CN110009405A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110555730A (en) * | 2019-08-28 | 2019-12-10 | 上海明品医学数据科技有限公司 | Data statistical analysis method for product after marketing research |
| CN111401941A (en) * | 2020-03-06 | 2020-07-10 | 武汉大学 | Vehicle sales prediction method based on XGboost recommendation algorithm |
| CN113554183A (en) * | 2021-08-03 | 2021-10-26 | 同济大学 | Extreme value prediction method based on unsupervised machine learning algorithm |
| CN113554183B (en) * | 2021-08-03 | 2022-05-13 | 同济大学 | Extreme value prediction method based on unsupervised machine learning algorithm |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110009400A (en) | Merchandise valuation method, terminal and computer readable storage medium | |
| CN110009405A (en) | A kind of passenger car sales volume simulating and predicting method based on network generalized extreme value model | |
| CN108647811A (en) | Predict that user buys method, apparatus, equipment and the storage medium of equity commodity | |
| JP2002513488A (en) | How to incorporate psychological effects into demand models | |
| US20140164220A1 (en) | Payment instrument selection | |
| CN107944593A (en) | A kind of resource allocation methods and device, electronic equipment | |
| CN110910180B (en) | Information pushing method and device, electronic equipment and storage medium | |
| US20110010211A1 (en) | Automatically prescribing total budget for marketing and sales resources and allocation across spending categories | |
| CN110111090A (en) | A kind of distribution method and device of electronics red packet | |
| CN105468628B (en) | A kind of sort method and device | |
| CN110162359A (en) | Method for pushing, the apparatus and system of new hand's guidance information | |
| CN109727055A (en) | Network Questionnaire Survey method, apparatus, equipment and computer readable storage medium | |
| CN113065882A (en) | Commodity processing method and device and electronic equipment | |
| CN109213936A (en) | product search method and device | |
| CN108038711A (en) | Customer loss Forecasting Methodology, device and computer equipment | |
| CN111626767B (en) | Resource data issuing method, device and equipment | |
| CN109492863A (en) | The automatic generation method and device of financial document | |
| Louviere | Modeling single individuals: the journey from psych lab to the app store | |
| CN107704974A (en) | Customer loss probability prediction algorithm and system under a kind of power sales | |
| Okudan et al. | Perceived feature utility-based product family design: a mobile phone case study | |
| CN109426983A (en) | Dodge purchase activity automatic generation method and device, storage medium, electronic equipment | |
| CN114298728A (en) | Data processing method and related device | |
| CN108510003A (en) | Car networking big data air control assemblage characteristic extracting method, device and storage medium | |
| US9798700B2 (en) | System and method for evaluating decisions using multiple dimensions | |
| CN115587238A (en) | Product brand marketing big data service method |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190712 |