Study On The Relevance Between Trip Characteristics And Real Estate Price Based On Mobile Signaling Data

Along with the economic growth in China,nowadays more and more people are living in the city urban area in China.The China's urbanization rate has jumped from 30% to53.4% in the past 20 years,which has caused housing inequality issues and serious traffic congestion.Generally,people living in the similar housing unit or district would have high degree of homogeneity in self attributes,e.g.social class,wealth,occupation,age.The different housing prices divide residents into different groups invisibly.And those groups of residents would have different travel behavior and daily activity patterns.We divide the housing area by the house average prices data into a number of residential districts to study people's trip characteristics in these districts.We mainly study the daily trip features of the different groups of residents,such as travel time,travel distance,trip origins and destinations,clustered by the housing information.Traditionally,residents' personal information and trip data are obtained by the manual investigations which cost quite a lot of manpower,time and money when the investigations are required to be carried out frequently and massively.Yet the cities in China are developing very fast,as well as the traffic demand.The traditional resident trip survey can no longer satisfy the frequent need of the urban planners' decision making.And the quality of the investigation results could be affected by subjective factors.Thus,it is vital to develop new method which can achieve the trip data continuously at an affordable cost and relatively high accuracy.Extracting resident trip data from mobile signaling data has become more and more popular in the past several years.The cellphone penetration rate is increasing at amazing speed,and most of people use phones everywhere and at any time.There are averagely 96.2 cellphones per one hundred of people in 2016 according to the government reports.The high cellphone penetration rate makes it possible to analyze urban traffic by using mobile signaling data.When subscribers use their cellphones,the mobile communication providers would get a large number of mobile signaling data including subscribers' ID,coordinates,time stamp.After data mining and filtering,every subscriber'sdaily trip chain can be roughly obtained.Dealing with the trip chain data,we can find out the subscriber's daily traffic information without the invasion of private information,including housing and oriental-destination locations,departure time,arrive time,travel time,travel distance,travel frequency,etc.This technology would be conducted regardless of the weather,manual factors and at an affordable cost.By using these mobile signaling data obtained from mobile communication providers,we can track millions of residents' traffic behaviors anonymously.This technology can help estimate the urban traffic demand constantly.Obviously,regional housing price and residents' trip pattern are highly correlated.People would choose their housing in consideration of traffic convenience,including the distance between their housing and workplace,the locations of subway or bus station,the hospitals,the Central Business District,etc.Meanwhile,the housing type,price and distribution also impact the residents' distribution,so as the traffic demand distribution.In this study,we combine the housing price information and resident trip information together and investigate the relationship between them.We focus on developing models to find out the trip characteristics of travelers clustered by their residential housing prices.