CN107093015A - A kind of analysis method of Car holding and usage behavior based on Sample Selection Model - Google Patents
A kind of analysis method of Car holding and usage behavior based on Sample Selection Model Download PDFInfo
- Publication number
- CN107093015A CN107093015A CN201710247917.5A CN201710247917A CN107093015A CN 107093015 A CN107093015 A CN 107093015A CN 201710247917 A CN201710247917 A CN 201710247917A CN 107093015 A CN107093015 A CN 107093015A
- Authority
- CN
- China
- Prior art keywords
- mrow
- car
- msub
- mtd
- built environment
- 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
- G06Q10/06—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
Abstract
The invention discloses the analysis method research of a kind of Car holding based on Sample Selection Model and usage behavior, and analyze residence and employment ground built environment institute role.Including 1, acquisition data：Mainly include extensive resident trip data, social economy's attribute, residence and place of working built environment data.2nd, the selection equation of Car holding is determined according to Sample Selection Model.3rd, second step is the result equation used the You Che colonies construction automobile chosen, and includes selection bias term.4th, model is built in statistical software, built environment is set as residence and employment ground built environment.5th, each parameter implicit realistic meaning and reason in analysis model result.The present invention considers endogenous and dependency relationships present in Car holding and use, using a kind of using Sample Selection Model as the conjunctive model relied on, and eliminates Sample Selection, the influence of research residence and employment ground built environment to car dependence.
Description
Technical field
The invention belongs to traffic behavior modeling and analysis field, specifically a kind of Sample Selection Model that is based on is to small vapour
Car possess and usage behavior analysis method research.
Background technology
Chinese urban population ratio rises to 54.77% in 2014 from 19.39% in 1980, is protected along with car
The rapid growth for the amount of having.From 2005 to 2014, China's car quantity increased nearly 10 times, has reached 106,000,000
.The thing followed is a series of traffic jam issue, problem of environmental pollution, problems of energy consumption.Government has to take
Some measures go to solve these problems, and restrict driving and limit purchase policy has become a big behave of government department pair, China it is big
City even some tier 2 cities all implement the policy that vehicle is restricted driving or vehicle limit is purchased.But people possess car or made
The anatomy being also worth with the influence factor of car behind, the reason for analyzing these deep layers contributes to traffic planners' planning more preferable
Urban structure, it is to avoid cause the Car holding of blindness and use.
Have scholar both at home and abroad to study the influence factor of Car holding, the shadow that some scholars use car
Sound is studied.But also both progress Conjoint Analysis of few people's joint, it is considered to the two dependence and endogeny for existing, draw
Enter to select bias term to eliminate Sample Selection against Mil Si (lambda).Whether present invention research commuter selects car to make
For trip instrument, thus residence built environment and employment ground built environment institute role are considered simultaneously.And obtained same
When considering residence with employment ground built environment, employment ground built environment and residence built environment are in Car holding and small
Automobile uses middle played the part of different role.
Compared to Car holding and the influence factor used is studied respectively in the past, the present invention combines modeling to both, right
Sample is selected, the commuter mode of research You Che colonies.Invention can eliminate the Select Error in both, be explained
Variable more accurately influence degree, secondly residence and considers duty residence relation is existed while employment ground built environment attribute
Role is brighter and clearer in commuting tools selection.
The content of the invention
The purpose of the present invention is to overcome the shortcomings of that existing investigative technique analyzes small vapour there is provided one kind based on Sample Selection Model
Car possess and usage behavior method.Research city is determined first, is carried out traffic survey data or is obtained from traffic programme department
Corresponding data are taken, the missing values and invalid data of the data obtained are handled.Secondly to samples selection in statistical software
Model is built, the samples selection equation of joint mapping Car holding, and the result equation that car is used.Foundation is built
Difference into environment attribute builds three group models respectively, is respectively set as residence built environment, employment ground built environment, and occupy
Residence and employment ground built environment.The influence journey that interpretation of result built environment finally according to model exists to Car holding
Degree, depth profiling is carried out according to Sample Selection to the endogeny and dependence between Car holding and use.If
The coefficient of built environment variable is significantly for just, then it represents that the built environment attribute has just to Car holding or the possibility used
To influence；Conversely, then there is negative sense influence.If the coefficient of inverse Mil Si is significantly, show there is Sample Selection, together
When against Mil's Si term coefficient significantly just to represent that unobservable factor has positive influence on Car holding when, while also to small vapour
Car is using there is positive influence.
The advantage of the invention is that：
(1) the characteristics of present invention is maximum is no longer individually to study the influence factor that Car holding and car are used, and
Consider the endogenous relation existed between the two, the samples selection of You Che colonies is carried out using a kind of method for combining modeling, is dissected
Duty residence built environment is to Car holding and the underlying causes used.
(2) in view of commuting pattern and duty residence built environment there is very big correlation in the present invention, thus in modeling
It is respectively residence built environment, employment ground built environment, and residence and employment ground built environment according to built environment attribute
Attribute builds three group models, is analyzed.
(3) eliminate due to Car holding present invention introduces inverse Mil Si of Sample Selection and using there is interior life
The Sample Selection of relation.
(4) present invention carries out instance analysis, obtains residence built environment to small according to the traffic survey data of Zhongshan city
The influence that automobile possesses is more than employment ground built environment attribute, and employment ground built environment attribute is more than to the influence that car is used
Residence built environment.
Brief description of the drawings
Fig. 1 is method flow schematic diagram of the invention；
Embodiment
Below in conjunction with accompanying drawing and example, the present invention is described in further detail.
The present invention provides a kind of analysis method of Car holding and usage behavior based on Sample Selection Model, flow chart
As shown in figure 1, comprising the following steps：
1), data acquisition and processing：Personal attribute, the data such as family's attribute and commuting tools selection need design traffic to adjust
Volume is interrogated, and provides and reclaims on a large scale, including the processing of shortage of data value and the processing of invalid questionnaire must all have strictly
Examination.Built environment attribute can be obtained from local traffic department, or according to investigator family and the GPS positions on employment ground
Put to calculate from map match software and obtain.
2) Sample Selection Model of Car holding, is set, it is ensured that there is at least one specific variable influence first step
The result of Car holding, and obvious influence is not produced on car commuting selection.Influence the variable of Car holding main
Including personal social economy's attribute and built environment attribute.First step model equation is as follows：
z* i=wiγ+biα+ε1i
probit(zi=1)=wiγ+biα+ε1i
ε1i~N (0,1)
Z=1 represents that individual possesses car, z in formula*It is a latent variable, w represents personal social economy's attribute, and γ is
The corresponding parameter of each variable, b is built environment attribute, and α is the corresponding parameter of each variable, ε1It is that Car holding was selected
Error term in journey, it is generally acknowledged that error term obedience standard is just distributed very much in Probit models.
3), equation selection is selected into the sample of result equation from first step Car holding, and to possessing car
Sample is modeled analysis, it is determined that there is car commuter to use the influence factor suffered by car.The explanation that car is used because
It is plain main to include personal social economy's attribute, built environment attribute, the selection bias term produced by selection course.Second
Portion's model equation is as follows：
The selection course of what above-mentioned formula showed be Car holding person, zi *What ＞ 0 was represented is that respondent has car, can be with
Its influence factor for using car is analyzed into second step.zi *What ＜ 0 was represented is that respondent does not have car, and second step is not entered and was analyzed
Journey.Whether second step analysis car owner is as follows with the Probit models of car：
probit(ci=1 | zi* ＞ 0)=xiβ+biξ+SB+ε2i
ε2i~N (0,1)
In formula, c=1 represents personal use car, and x is the built environment attribute that result equation needs to consider, β is correspondence
Parameter, b represents built environment attribute, and ξ is corresponding parameter, and SB is the bias term produced during samples selection, ε2It is knot
The error term of fruit equation, thinks obedience standard just too in Probit models.Neutralized in conjunctive model and think two error clothes
From standard, two are just being distributed very much, and as follows, ρ is the coefficient correlation of two error term：
4), also referred to as inverse Mil Si of Sample Selection, its generation is due to that Car holding is not one random
The process of distribution, and it is affected by various factors, including Observable and unobservable factor.Observable factor includes can root
Social economy's attribute that questionnaire is obtained according to investigations, and obtained built environment attribute is calculated, unobservable factor is then gone through
History cultural traits, and regional custom custom etc..And select bias term to be then that one produced during nonrandom assume responsibility for something lost
A variable of variable effect is leaked, selection bias term is introduced in second step result equation, it is estimated that more accurately mould
Type result.
In formula, φ () is the probability density function of standardized normal distribution, and Φ () is the cumulative distribution letter of standardized normal distribution
Number.
5) meaning and reason of built environment attributes estimation result, the underlying causes that analysis selection deviation implies, are analyzed.
α ＞ 0 represent that built environment variable produces positive facilitation to Car holding, and α ＜ 0 represent that built environment is gathered around to car
There is the reduction effect for producing negative sense.ξ ＞ 0 represent that built environment attribute uses the facilitation for producing forward direction, ξ ＜ 0 to car
Represent that built environment attribute is acted on using the reduction for producing negative sense car.ρ ＞ 0 represent that Car holding and car are used
There is positive correlation in the error term in two step equations, when observable factor does not produce positive acting to Car holding, simultaneously
Also generation positive acting is used to car.
Embodiment
1), choose exemplified by Zhongshan city's traffic survey data, the traffic survey data of Zhongshan city, choosing are imported in statistical software
Explanatory variable is taken, model construction is carried out.The present embodiment is only provided while introducing residence and the model of built environment attribute variable
As a result, the modeling process for being introduced separately into residence and built environment attribute is consistent with embodiment.What constructed two steps model was introduced
Explanatory variable is as follows：
The explanatory variable of Car holding includes：Whether family's child's number, family has passport, and family whole year total income is less than 2
Ten thousand, family whole year total income is more than 60,000, H job density, and H_ intersections density, H_ bus stations density, H_ soils mixing makes
With entropy, H_CBD distances, W_ job density, W_ intersections density, W_ bus stations density, W_ mixed-use districts entropys, W_
CBD distances.Whether wherein family child number, family has passport, and family whole year total income is less than 20,000, and family whole year total income is more than 60,000
It is that larger peculiar variable is influenceed on family's buying car Deng personal social property.The variable of H_ beginnings represents the built environment of residence
Attribute, the variable of W_ beginnings represents the built environment attribute on employment ground.
The explanatory variable that car is used includes：Sex, age 18-30, the age is more than 60, civil servant, kinsfolk
Number, H_ job density, H_ intersections density, H_ bus stations density, H_ mixed-use districts entropys, H_CBD distances, W_ is just
Industry post density, W_ intersections density, W_ bus stations density, W_ mixed-use districts entropy W_CBD distances.
2) due to often there is correlation between the explanatory variable of model result, the conspicuousness of model estimated result is caused
There is deviation, the embodiment carries out multicollinearity detection to built environment variable, testing result is as shown in the table.Each variable
Variance inflation factor (Variance Inflation Factor, VIF) value be both less than 5, it is believed that selected builds up ring
Multicollinearity is not present between the attribute of border.
| Built environment variable | VIF |
| H_ job density | 2.5 |
| H_ intersections density | 2.43 |
| W_ intersections density | 2.29 |
| W_ job density | 2.23 |
| W_ bus stations density | 2.17 |
| H_CBD distances | 2.15 |
| W_CBD distances | 2.11 |
| H_ bus stations density | 2.08 |
| H_ mixed-use districts entropys | 1.52 |
| W_ mixed-use districts entropys | 1.51 |
| Average VIF values | 2.1 |
3) model estimated result.Following table is the model result for possessing car, and personal society is understood from selection equation result
Meeting economic attribution influences bigger than built environment attribute to possessing for car.The built environment attribute of residence is to Car holding
Influence be more than employment ground built environment.
| Car holding | Estimation coefficient | Standard deviation | Z values | P>z |
| Family's child's number | 0.099 | 0.009 | 10.450 | 0.000 |
| Whether family has driving license | 1.621 | 0.025 | 63.910 | 0.000 |
| Family whole year total income is less than 20,000 | -0.410 | 0.037 | -11.150 | 0.000 |
| Family whole year total income is more than 60,000 | 0.961 | 0.019 | 51.800 | 0.000 |
| H_ job density | -0.008 | 0.003 | -2.490 | 0.013 |
| H_ intersections density | 0.008 | 0.002 | 3.780 | 0.000 |
| H_ bus stations density | 0.029 | 0.006 | 4.880 | 0.000 |
| H_ mixed-use districts entropys | 0.181 | 0.060 | 3.040 | 0.002 |
| H_CBD distances | -0.002 | 0.001 | -1.490 | 0.136 |
| W_ job density | -0.002 | 0.003 | -0.810 | 0.421 |
| W_ intersections density | 0.001 | 0.002 | 0.460 | 0.644 |
| W_ bus stations density | 0.017 | 0.006 | 3.030 | 0.002 |
| W_ mixed-use districts entropys | -0.136 | 0.061 | -2.220 | 0.026 |
| W_CBD distances | 0.004 | 0.001 | 3.530 | 0.000 |
| Constant | -2.472 | 0.054 | -45.640 | 0.000 |
From following table it can be seen that in control Sample Selection against after Mil Si, personal social economy's attribute makes to car
Influence is substantially bigger than built environment attribute, and this is the same with expected result.But unlike that Car holding knot
Fruit, the influence that employment ground built environment attribute is used car is more than residence built environment.And inverse Mil Si is than number
Be negative indication when unobservable factor on Car holding forward direction influence when, have negative sense influence to the use of car on the contrary.
This is related to the habits and customs of China resident.(such as job density, intersection is close for the factor of some promotion Car holdings
Degree), often due to it causes congestion etc. to be not easy to the situation of trip, the influence commuted in turn to car is negative role.
| Use car | Estimation coefficient | Standard deviation | Z values | P>z |
| Sex | 1.227 | 0.033 | 37.640 | 0.000 |
| Age 18-30 | -0.398 | 0.035 | -11.500 | 0.000 |
| Age>60 | -1.165 | 0.137 | -8.510 | 0.000 |
| Civil servant | 0.428 | 0.055 | 7.820 | 0.000 |
| Kinsfolk's number | -0.258 | 0.014 | -18.670 | 0.000 |
| H_ job density | -0.008 | 0.006 | -1.390 | 0.165 |
| H_ intersections density | -0.003 | 0.004 | -0.680 | 0.495 |
| H_ bus stations density | 0.024 | 0.010 | 2.410 | 0.016 |
| H_ mixed-use districts entropys | -0.120 | 0.107 | -1.120 | 0.262 |
| H_CBD distances | -0.001 | 0.002 | -0.310 | 0.753 |
| W_ job density | 0.003 | 0.004 | 0.720 | 0.473 |
| W_ intersections density | -0.009 | 0.004 | -2.610 | 0.009 |
| W_ bus stations density | 0.032 | 0.009 | 3.420 | 0.001 |
| W_ mixed-use districts entropys | 0.190 | 0.107 | 1.770 | 0.077 |
| W_CBD distances | 0.003 | 0.002 | 1.240 | 0.216 |
| Inverse Mil Si | -0.617 | 0.039 | -15.660 | 0.000 |
| Constant | 0.689 | 0.117 | 5.880 | 0.000 |
4) model result is understood：
The embodiment is with Zhongshan city's traffic survey data and built environment data instance, it was demonstrated that Car holding and small vapour
A kind of endogenous relation is there is between car use.Simultaneously after the personal social economy's attribute of control, built environment is still
To possessing and using there is significant impact for car.Wherein influence of the built environment attribute in residence to Car holding is more than
Employment ground；The influence that employment ground built environment attribute is used car is more than residence.This shows for there is the commuter of car
For, the built environment of residence plays more importantly role in car in.
It is described in detail above it is of the invention be preferable to carry out case, but the invention is not limited in above-mentioned case study on implementation
The part steps of the present invention in the range of the overall structure of the present invention, can be carried out a variety of conversion and again group by detail
Close, the present invention is no longer enumerated to various possible combinations, these conversion combinations belong to protection scope of the present invention.
Claims (5)
1. the present invention provides a kind of analysis method of Car holding and usage behavior based on Sample Selection Model, flow chart is such as
Shown in Fig. 1, comprise the following steps：
1), data acquisition and processing：Personal attribute, the data such as family's attribute and commuting tools selection need design traffic study to ask
Volume, and provide and reclaim on a large scale, including the processing of shortage of data value and the processing of invalid questionnaire must all have strict examine
Look into.Built environment attribute can be obtained from local traffic department, or according to the GPS location on investigator family and employment ground from
Calculate and obtain in map match software.
2) Sample Selection Model of Car holding, is set, it is ensured that there is at least one specific variable influence small vapour of the first step
The result that car possesses, and obvious influence is not produced on car commuting selection.The variable of influence Car holding mainly includes
Personal social economy's attribute and built environment attribute.First step model equation is as follows：
z* i=wiγ+biα+ε1i
probit(zi=1)=wiγ+biα+ε1i
ε1i~N (0,1)
Z=1 represents that individual possesses car, z in formula*It is a latent variable, w represents personal social economy's attribute, and γ is each change
Corresponding parameter is measured, b is built environment attribute, and α is the corresponding parameter of each variable, ε1In being Car holding selection course
Error term, it is generally acknowledged that error term obedience standard is just distributed very much in Probit models.
3), equation selection is selected into the sample of result equation from first step Car holding, and to possessing the sample of car
Analysis is modeled, it is determined that there is car commuter to use the influence factor suffered by car.The explanation factor master that car is used
To include personal social economy's attribute, built environment attribute, the selection bias term produced by selection course.Second mould
Type equation is as follows：
<mrow>
<msub>
<mi>z</mi>
<mi>i</mi>
</msub>
<mi>i</mi>
<mi>s</mi>
<mfenced open = "{" close = "">
<mtable>
<mtr>
<mtd>
<mrow>
<mi>o</mi>
<mi>b</mi>
<mi>s</mi>
<mi>e</mi>
<mi>r</mi>
<mi>v</mi>
<mi>e</mi>
<mi>d</mi>
</mrow>
</mtd>
<mtd>
<mtable>
<mtr>
<mtd>
<mrow>
<mi>i</mi>
<mi>f</mi>
</mrow>
</mtd>
<mtd>
<mrow>
<msup>
<msub>
<mi>z</mi>
<mi>i</mi>
</msub>
<mo>*</mo>
</msup>
<mo>></mo>
<mn>0</mn>
</mrow>
</mtd>
</mtr>
</mtable>
</mtd>
</mtr>
<mtr>
<mtd>
<mrow>
<mi>u</mi>
<mi>n</mi>
<mi>o</mi>
<mi>b</mi>
<mi>s</mi>
<mi>e</mi>
<mi>r</mi>
<mi>v</mi>
<mi>e</mi>
<mi>d</mi>
</mrow>
</mtd>
<mtd>
<mtable>
<mtr>
<mtd>
<mrow>
<mi>i</mi>
<mi>f</mi>
</mrow>
</mtd>
<mtd>
<mrow>
<msup>
<msub>
<mi>z</mi>
<mi>i</mi>
</msub>
<mo>*</mo>
</msup>
<mo><</mo>
<mn>0</mn>
</mrow>
</mtd>
</mtr>
</mtable>
</mtd>
</mtr>
</mtable>
</mfenced>
</mrow>
The selection course of what above-mentioned formula showed be Car holding person, zi *What ＞ 0 was represented is that respondent has car, can be entered
Second step analyzes the influence factor that it uses car.zi *What ＜ 0 was represented is that respondent does not have car, and second step analysis process is not entered.
Whether second step analysis car owner is as follows with the Probit models of car：
probit(ci=1 | zi *＞ 0)=xiβ+biξ+SB+ε2i
ε2i~N (0,1)
In formula, c=1 represents personal use car, and x is the built environment attribute that result equation needs to consider, β is corresponding ginseng
Number, b represents built environment attribute, and ξ is corresponding parameter, and SB is the bias term produced during samples selection, ε2It is result side
The error term of journey, thinks obedience standard just too in Probit models.Neutralized in conjunctive model and think that two error obey are marked
Accurate two are just distributed very much, and as follows, ρ is the coefficient correlation of two error term：
<mrow>
<mo>(</mo>
<msub>
<mi>&epsiv;</mi>
<mrow>
<mn>1</mn>
<mi>i</mi>
</mrow>
</msub>
<mo>,</mo>
<msub>
<mi>&epsiv;</mi>
<mrow>
<mn>2</mn>
<mi>i</mi>
</mrow>
</msub>
<mo>)</mo>
<mo>~</mo>
<mi>N</mi>
<mo>(</mo>
<mrow>
<mfenced open = "(" close = ")">
<mtable>
<mtr>
<mtd>
<mn>0</mn>
</mtd>
</mtr>
<mtr>
<mtd>
<mn>0</mn>
</mtd>
</mtr>
</mtable>
</mfenced>
<mo>,</mo>
<mfenced open = "(" close = ")">
<mtable>
<mtr>
<mtd>
<mn>1</mn>
</mtd>
<mtd>
<mi>&rho;</mi>
</mtd>
</mtr>
<mtr>
<mtd>
<mi>&rho;</mi>
</mtd>
<mtd>
<mn>1</mn>
</mtd>
</mtr>
</mtable>
</mfenced>
</mrow>
<mo>)</mo>
</mrow>
4), also referred to as inverse Mil Si of Sample Selection, its generation is due to that Car holding is not one and is randomly assigned
Process, it is and affected by various factors, including Observable and unobservable factor.Observable factor includes can be according to tune
Social economy's attribute that volume is obtained is interrogated, and calculates obtained built environment attribute, unobservable factor then has history text
Change feature, and regional custom custom etc..And select bias term to be then that one produced during nonrandom assume responsibility for omitting change
One variable of amount effect, introduces selection bias term in second step result equation, it is estimated that more accurately model knot
Really.
<mrow>
<mi>S</mi>
<mi>B</mi>
<mo>=</mo>
<mi>&rho;</mi>
<mfrac>
<mrow>
<mi>&phi;</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>w</mi>
<mi>i</mi>
</msub>
<mi>&gamma;</mi>
<mo>+</mo>
<msub>
<mi>b</mi>
<mi>i</mi>
</msub>
<mi>&alpha;</mi>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<mi>&Phi;</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>w</mi>
<mi>i</mi>
</msub>
<mi>&gamma;</mi>
<mo>+</mo>
<msub>
<mi>b</mi>
<mi>i</mi>
</msub>
<mi>&alpha;</mi>
<mo>)</mo>
</mrow>
</mrow>
</mfrac>
</mrow>
<mfenced open = "" close = "">
<mtable>
<mtr>
<mtd>
<mrow>
<mi>p</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>c</mi>
<mi>i</mi>
</msub>
<mo>=</mo>
<mn>1</mn>
<mo>|</mo>
<msup>
<mi>z</mi>
<mo>*</mo>
</msup>
<mo>></mo>
<mn>0</mn>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mi>p</mi>
<mi>r</mi>
<mi>o</mi>
<mi>b</mi>
<mi>i</mi>
<mi>t</mi>
<mo>&lsqb;</mo>
<mi>E</mi>
<mrow>
<mo>(</mo>
<msup>
<mi>c</mi>
<mo>*</mo>
</msup>
<mo>></mo>
<mn>0</mn>
<mo>|</mo>
<msup>
<mi>z</mi>
<mo>*</mo>
</msup>
<mo>></mo>
<mn>0</mn>
<mo>)</mo>
</mrow>
<mo>&rsqb;</mo>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mrow>
<mo>=</mo>
<mi>&Phi;</mi>
<mo>&lsqb;</mo>
<msub>
<mi>x</mi>
<mi>i</mi>
</msub>
<mi>&beta;</mi>
<mo>+</mo>
<msub>
<mi>b</mi>
<mi>i</mi>
</msub>
<mi>&alpha;</mi>
<mo>+</mo>
<mi>&rho;</mi>
<mfrac>
<mrow>
<mi>&phi;</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>w</mi>
<mi>i</mi>
</msub>
<mi>&gamma;</mi>
<mo>+</mo>
<msub>
<mi>b</mi>
<mi>i</mi>
</msub>
<mi>&alpha;</mi>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<mi>&Phi;</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>w</mi>
<mi>i</mi>
</msub>
<mi>&gamma;</mi>
<mo>+</mo>
<msub>
<mi>b</mi>
<mi>i</mi>
</msub>
<mi>&alpha;</mi>
<mo>)</mo>
</mrow>
</mrow>
</mfrac>
<mo>&rsqb;</mo>
</mrow>
</mtd>
</mtr>
</mtable>
</mfenced>
In formula, φ () is the probability density function of standardized normal distribution, and Φ () is the Cumulative Distribution Function of standardized normal distribution.
5) meaning and reason of built environment attributes estimation result, the underlying causes that analysis selection deviation implies, are analyzed.α ＞ 0
Represent that built environment variable produces positive facilitation to Car holding, α ＜ 0 represent that built environment is produced to Car holding
The reduction effect of raw negative sense.ξ ＞ 0 represent that built environment attribute is represented using positive facilitation, ξ ＜ 0 is produced car
Built environment attribute is acted on using the reduction for producing negative sense car.ρ ＞ 0 represent that Car holding and car use two steps
There is positive correlation in the error term in equation, when observable factor does not produce positive acting to Car holding, while also right
Car uses generation positive acting.
2. the analysis method of a kind of Car holding and usage behavior based on Sample Selection Model according to claim 1
Step 3) in, based on the continuous results model of Heckman models, it is proposed that the binary outcome model that car is used.
3. the analysis method of a kind of Car holding and usage behavior based on Sample Selection Model according to claim 1
Step 2) and step 3) in built environment attribute not only include the built environment attribute of residence, in addition to employment ground is built up
Environment attribute.
4. the analysis method of a kind of Car holding and usage behavior based on Sample Selection Model according to claim 1
Step 4) in, select binary inverse Mil Si of joint binary outcome variate model to carry out the derivation of theoretical formula than number, and
The deciphering of practical significance.
5. draw 1) residence built environment category to having carried out instance analysis using Zhongshan city's traffic survey data of 2010
Property influence to Car holding be more than employment ground built environment attribute；2) employment ground built environment attribute commutes to You Che colonies
It is more than residence built environment attribute using the influence of car.And learn that unobservable factor has forward direction to possessing car
During influence, on there is negative sense influence on the contrary using car.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710247917.5A CN107093015A (en) | 2017-04-17 | 2017-04-17 | A kind of analysis method of Car holding and usage behavior based on Sample Selection Model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710247917.5A CN107093015A (en) | 2017-04-17 | 2017-04-17 | A kind of analysis method of Car holding and usage behavior based on Sample Selection Model |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107093015A true CN107093015A (en) | 2017-08-25 |
Family
ID=59638248
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710247917.5A Pending CN107093015A (en) | 2017-04-17 | 2017-04-17 | A kind of analysis method of Car holding and usage behavior based on Sample Selection Model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107093015A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110851670A (en) * | 2019-10-25 | 2020-02-28 | 袁茂银 | Underground cable fault repairing method and device |
| CN112700283A (en) * | 2021-01-06 | 2021-04-23 | 东南大学 | Method for improving shared automobile utilization rate based on latent variable |
-
2017
- 2017-04-17 CN CN201710247917.5A patent/CN107093015A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110851670A (en) * | 2019-10-25 | 2020-02-28 | 袁茂银 | Underground cable fault repairing method and device |
| CN112700283A (en) * | 2021-01-06 | 2021-04-23 | 东南大学 | Method for improving shared automobile utilization rate based on latent variable |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Arora et al. | Comparative evaluation of geospatial scenario-based land change simulation models using landscape metrics | |
| Van Sickle | Using mean similarity dendrograms to evaluate classifications | |
| CN103995952B (en) | The Mining Wasteland of a kind of improvement is reclaimed suitability fuzzy synthetic appraisement method | |
| Kovács et al. | Classification into homogeneous groups using combined cluster and discriminant analysis | |
| CN104217002B (en) | A kind of road condition information filling method obtained based on quality data | |
| CN111898315B (en) | Landslide susceptibility assessment method based on fractal-machine learning hybrid model | |
| CN102521624B (en) | Classification method for land use types and system | |
| CN106228808A (en) | City expressway travel time prediction method based on Floating Car space-time grid data | |
| CN102184423B (en) | Full-automatic method for precisely extracting regional impervious surface remote sensing information | |
| CN104200104A (en) | Fine granularity air pollutant concentration area estimation method based on spatial characteristics | |
| Jarosław et al. | Using multivariate statistics and fuzzy logic system to analyse settlement preferences in lowland areas of the temperate zone: an example from the Polish Lowlands | |
| He et al. | A catchment classification scheme using local variance reduction method | |
| CN107093015A (en) | A kind of analysis method of Car holding and usage behavior based on Sample Selection Model | |
| CN110070255A (en) | The method for introducing commuter's Passenger Traveling Choice modeling and analysis after sharing bicycle | |
| CN109118004A (en) | A kind of engineer construction addressing Suitable Area prediction technique | |
| Zhang et al. | An atmospheric vulnerability assessment framework for environment management and protection based on CAMx | |
| Liu et al. | A comparative evaluation of machine learning algorithms and an improved optimal model for landslide susceptibility: a case study | |
| Chen et al. | The application of the genetic adaptive neural network in landslide disaster assessment | |
| CN106485716A (en) | A kind of many regarding SAR image segmentation method with Gamma mixed model based on region division | |
| Rahimabadi et al. | An ecological agricultural model using fuzzy AHP and PROMETHEE II approach. | |
| CN113792992B (en) | Landslide collapse emergency evaluation method and system based on remote sensing big data | |
| CN108053053A (en) | The non-commuting travel forecasting approaches of electric bicycle based on space autoregression model | |
| Nasiri et al. | Implementation of agricultural ecological capability model using integrated approach of PROMETHEE II and fuzzy-AHP in GIS environment (Case Study: Marvdasht county) | |
| CN109523143B (en) | Land evaluation method based on multi-granularity calculation | |
| CN108182170A (en) | Electric bicycle commuter Forecasting Methodology based on space autoregression preference pattern |
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: 20170825 |