Method for accurately measuring overflow range of urban real estate market based on feature price model

Abstract

The invention discloses a method for accurately measuring the overflow range of an urban real estate market based on a feature price model, and the method comprises the following steps: collecting and researching housing price data of a main built-up area range of a city, and obtaining a housing price point data set; performing kernel density estimation according to the housing price point data set to generate a polygonal cluster set; calculating an urban cross-boundary space continuous area according to the polygon cluster set; and calculating the housing economic overflow area according to the feature price model and the city cross-boundary space continuous area. According to the method, the overflow effect is measured through the real estate market economy visual angle, the remodeled urban boundary space geographic features are recognized, and a new visual angle and method can be provided for understanding a real estate market economy overflow operation mechanism and policy setting.

Description

Technical Field

[0001] The present invention relates to the field of real estate market structure research, and in particular to a method for accurately measuring the spillover range of urban real estate markets based on a characteristic price model. Background Art

[0002] As a widely recognized economic phenomenon, the spillover effect of the real estate market is mostly used to describe the interaction between cities or urban areas at the macro level, but there is a lack of spatial description of spillover phenomena at a finer scale. The traditional measurement of the spillover effect of the real estate market focuses on the causal correlation and impact intensity measurement in the field of econometrics, and is usually based on house price time series data with large data volume, high collection difficulty and slow update speed. The results presented in data tables or network diagrams are not very readable and can hardly play an effective supporting role in practical application and planning decision-making.

[0003] The hedonic price model is a classic tool for evaluating, predicting and analyzing influencing factors of real estate market prices. It is often used in price index formulation and sub-market division, but its connection with space is relatively weak and the description of heterogeneity of real estate markets between cities is insufficient. At present, the measurement of urban real estate market spillovers is mostly based on the economic perspective of ripple effect and bubble effect, which stays at the macro-impact between cities and is difficult to go deep into the district and county or even street group scale. If the optimal fit of the hedonic price model can be used as the analysis scale of the heterogeneity measurement index, it is free and flexible, and the cross-border spillover space range of the urban real estate market economy can be solved more accurately, which is of great significance for understanding the influencing factors and transmission mechanism of spillover effects. However, no relevant research and application have been seen so far. Summary of the invention

[0004] In view of the above-mentioned problems existing in the prior art, the present invention provides a method for accurately measuring the spillover scope of the urban real estate market based on the characteristic price model. By measuring the spillover effect from the perspective of the real estate market economy and re-recognizing the spatial geographical characteristics of the urban boundary under its reshaping, a new perspective and method can be provided for understanding the operating mechanism of the real estate market economic spillover and the implementation of policies.

[0005] The technical solution of the present invention is as follows: A method for accurately measuring the spillover range of urban real estate market based on a characteristic price model includes the following steps: S1. Collect spatial vector data of housing prices in the main built-up areas of the research city, define them as housing price sample points, and obtain a set of housing price sample points, i.e., housing price sample point set HP; the housing price sample points include: housing unit price information, housing structure information, and housing latitude and longitude information; the housing structure information includes: area, number of bedrooms, number of bathrooms, and floors; S2. Based on the housing price sample point set HP, the “kernel density estimation” tool of ArcGIS pro is used to generate a raster dataset, which is recorded as Grid_KDE. S3. Based on the raster dataset Grid_KDE, use the “Select by Attribute” tool of ArcGIS Pro to generate a set of polygon clusters, which is recorded as the urban natural residential area cluster set G; S4. Set the spatial proximity distance threshold W, and traverse each polygon in the urban natural residential area cluster set G according to the following rules: for the selected polygon i, calculate its boundary distance to other polygons in G and obtain the minimum value, which is recorded as Di. If Di≤W, mark the polygon; the boundary distance is the shortest distance between the boundaries of two polygons; after the traversal, all marked polygons are combined into a new set, which is the theoretical urban cross-border spatial continuous area, recorded as SSOA; S5. Select the initial city A, whose housing price sample point set is HP A According to HP A Calculate the hedonic price model M A , and obtain the initial goodness of fit R 2 A In the SSOA area of ​​the city, the housing price sample points HP corresponding to each polygon cluster i in the urban natural residential area cluster set G are arranged according to the principle of centroid distance from near to far. i Add to HP one by one A Use M again A Calculate and exclude polygonal clusters whose significant factor Beta is reversed after inclusion, and retain the goodness of fit R after inclusion 2 A The rising polygon clusters, until all SSOA regions are traversed to reach R 2 A The optimal, remaining polygonal clusters are the homogeneous housing economic spillover areas that can improve the explanatory power of the characteristic price model, denoted as HSSOA; S6. First, calculate the Euclidean distance X between each polygon cluster in HSSOA and the five important geographical features of the initial city A: medical, transportation, education, commerce, and landscape. H , X T , X E , X S , X L , and then add these five Euclidean distances as new dependent variables into the hedonic price model M A The EHSSOA of the region is further calculated based on the factor significance and the exogenous factors with key driving force are further extracted.

[0006] Furthermore, the specific steps of S3 are as follows: S3-1. Take Grid_KDE as the input grid, use the "Reclassify" tool of ArcGIS Pro, set the classification method to "Natural Breaks Method", keep the number of categories as default, and run to generate a grid dataset, recorded as Grid_Reclass; compared with Grid_KDE with continuous floating point density values, the grid value of Grid_Reclass is a positive integer between 1 and 5 under default conditions; S3-2, using Grid_Reclass as input grid, use ArcGIS Pro's "Select by Attribute" tool to select grids with values ​​greater than 1, then use ArcGIS Pro's "Raster to Polygon" tool to select grids with values ​​greater than 1 as input grids, uncheck the "Simplify Polygon" option, run to generate polygon datasets and export them as shapefiles. The polygons are the natural residential clusters in the city, recorded as Shp_Cluster; S3-3. Take Shp_Cluster as input data, use the "Feature to Point" tool of ArcGIS Pro, check the "Internal Point" option, run to generate the cluster centroid point dataset, and export the shapefile file, recorded as Shp_Center.

[0007] Furthermore, the specific steps of S4 are as follows: S4-1. Take Shp_Cluster as input data, use the "Pairwise Buffer" tool of ArcGIS Pro, set the buffer distance to the spatial proximity distance threshold W, uncheck "Export all features as a single feature surface", run to generate a polygon dataset and export the shapefile file, recorded as Shp_Buffer; S4-2, the selected city is recorded as SC, the set of its adjacent cities is recorded as AC, the shapefiles storing the polygonal surfaces of the administrative boundaries of the above two cities are recorded as Shp_SC and Shp_AC, and the shapefiles storing the sample point data of the housing prices of the above two cities are recorded as HP_SC and HP_AC; S4-3. Use Python to calculate the SSOA of the cross-boundary spatial contiguous area of ​​the city. The specific steps are as follows: S4-3-1, call the geopandas library, read and select all points in Shp_Center that are in Shp_SC, and record this point set as SC_Center; S4-3-2, call the geopandas library, read and select all polygons containing SC_Center in Shp_Buffer, and record the polygon set as SC_Buffer; S4-3-3, call the geopandas library, read and select all polygons in Shp_Cluster that intersect with SC_Buffer, and record the polygon set as SC_Cluster; S4-3-4. Call the geopandas library to read and select all polygons in SC_Cluster that are in Shp_AC. This polygon set is the cross-boundary spatial continuous area SSOA of SC.

[0008] Furthermore, in step S5, Python is used to calculate the urban cross-border housing spillover area HSSOA, and the specific steps are as follows: S5-1. Call the geopandas library to calculate the Euclidean distance between each polygon in the cross-border spatial contiguous area SSOA and the SC boundary, and re-sort them in ascending order; S5-2, set the initial value HP_INI of the housing price sample point set HP to HP_SC; read the polygons P in SSOA one by one in the order determined in S5-1; each time a polygon P is read, perform the operations from S5-2-1 to S5-2-4: S5-2-1, read and select all points in HP that are in polygon P, and merge them with HP_INI to obtain HP_Merge; S5-2-2. Establish a hedonic price model for HP_Merge, using variables including house structure information, location characteristics and facility distance; the house structure information includes the area, number of bedrooms, number of bathrooms and floors in data collection; the location characteristics include road selectivity and integration calculated using the spatial syntax software Depthmap; the facility distance includes the distance to hospitals, primary schools, middle schools, supermarkets and other facilities; S5-2-3. Set a threshold for VIF, namely, VIF threshold; VIF is variance inflation factor; call the statsmodels library in Python to use the variance_inflation_factor function to calculate multicollinearity for all variables, and gradually remove the highest factor to calculate VIF again until the VIF of all variables is less than or equal to the VIF threshold, and obtain the variables for constructing the final model; S5-2-4. Call the statsmodels library in Python to use the OLS function to perform multiple linear regression and obtain the goodness of fit R of HP_INI and HP_Merge respectively. 1 2 and R 2 2 ; if R 2 2 >R1 2 , then record the polygon P read this time and update HP_INI to HP_Merge; otherwise, do not update; S5-3. After traversing all polygons in SSOA according to the above steps, all recorded polygons P that meet the conditions are defined as a set. This polygon data set is the housing spillover area HSSOA of SC.

[0009] Furthermore, in step S6, Python is used to calculate the EHSSOA of the cross-border external source impact area of ​​the city. The specific steps are as follows: S6-1, read the five types of POI point data sets located in SC, namely "hub site", "school", "tertiary hospital", "shopping center" and "scenic spot", and record them as POI_Tran, POI_Edu, POI_Hos, POI_Com and POI_Lan respectively; S6-2, read the polygons in HSSOA one by one, and execute the operations of S6-2-1 to S6-2-4 every time a polygon P is read: S6-2-1. Read and select the point in Shp_Center located in polygon P, calculate the distance from the point to POI_Tran, POI_Edu, POI_Hos, POI_Com, POI_Lan, and select the three closest distances D 1 , D 2 , D 3 And use it as the new distance variable group X add ; S6-2-2, read and select all points in HP that are in polygon P, record them as HP_Sel; establish a characteristic price model for HP_Sel, using variables including the distance variable group calculated in the previous step, house structure information, location characteristics and facility distance; S6-2-3. Call the statsmodels library in Python to use the variance_inflation_factor function to calculate multicollinearity for all variables, and gradually remove the factors with the highest VIF and calculate the VIF again until the VIF of all variables is less than or equal to the VIF threshold, and finally obtain the variables for constructing the model; S6-2-4. Call the statsmodels library in Python to use the OLS function to perform multiple linear regression, obtain the fitting result of HP_Sel, and determine the newly added distance variable X add Is it significant? If so, define the polygon P as affected by X. add The external source of influence area driven is recorded, otherwise it is not recorded; S6-3. After traversing all polygons in HSSOA according to the above steps, all recorded polygons that meet the conditions are included in a set. This polygon data set is the external source influence area EHSSOA of SC.

[0010] Furthermore, the spatial proximity distance threshold W of step S4 is 6000; The VIF threshold in step S5-2-3 and step S6-2-3 is 5; In step S6-2-4, the distance variable X is determined. add The condition for significance is: head-to-tail split ratio p<0.05.

[0011] The beneficial technical effects of the present invention are: The present invention proposes a method based on a characteristic price model driven by multi-source data, which is applicable to solving the spatial scope of spillover of real estate market in any city, and is deep-rooted in the district and county and even street group scale, and has strong scalability; The multi-source data used in the present invention include urban road network data, point of interest data and real estate price data (including spatial location and unit information). The data is highly open and easy to obtain, does not require the time continuity of the data, and can be applied to any time section of the city. The calculation results obtained by the present invention not only include traditional spillover impact data tables and network diagrams, but also directly provide a spatial range map of urban real estate market spillovers, and further divide the spillover space into three layers from weak to strong: spatial continuous areas, housing spillover areas, and exogenous influence areas, and obtain empirical results of circle-type spillovers; The spillover space measurement results of the present invention can be accurate to the scale of residential clusters of 100 meters, and deeply reveal its internal driving factors and action mechanisms, helping decision makers to re-recognize the geographical characteristics of urban space reshaped by spillover effects, and provide a new perspective and reliable methods for planning implementation and policy implementation. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 It is a logic block diagram of the present invention. DETAILED DESCRIPTION

[0013] The present invention is described in detail below in conjunction with the accompanying drawings and embodiments. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0014] The embodiment adopts the present invention to identify the structure of real estate market spillover in Shanghai. The specific steps are as follows: Figure 1 As shown: S1. Collect housing price data in the Yangtze River Delta region by crawling second-hand housing transaction websites such as Anjuke and Beike (or other data acquisition methods), including house unit price information, house structural attributes such as area, number of bedrooms, number of bathrooms, and floors to facilitate the subsequent establishment of a characteristic price model, as well as longitude and latitude information for geocoding into shapefile spatial data.

[0015] S2. Correct the housing price data to obtain data in the WGS1984 coordinate system, and use the spatial location filtering tool in ArcGIS Pro to filter out the housing price data using the administrative division data of the Yangtze River Delta region (also in the WGS1984 coordinate system) to eliminate the location error caused by geocoding. The filtered housing price data is defined as housing price sample points, and the set of housing price sample points is HP.

[0016] S3. Use ArcGIS Pro to perform spatial processing on the housing price sample point set HP. The specific steps are as follows: S3-1. Input the xy coordinates of the housing price sample point set HP into ArcGIS Pro to obtain the housing price point dataset, and use the "Project" tool to change the projection coordinate system of the dataset to a suitable coordinate system (for national research, it can be projected to the Lambert coordinate system, and for Shanghai research, it can be projected to the WGS_1984_51N coordinate system), and export it to a shapefile file denoted as Shp_Pt.

[0017] S3-2. Take the housing price point dataset Shp_Pt as input data, use the ArcGIS pro "Kernel Density Estimate" tool, set the search radius (bandwidth) to r=1000, the unit is meter (m), and the grid size to any integer value between 10 and 100m (the smaller the value, the higher the accuracy), and run the generated raster dataset, recorded as Grid_KDE.

[0018] S3-3. Take Grid_KDE as the input raster, use the "Reclassify" tool of ArcGIS Pro, set the classification method to "Natural Breaks Method", keep the number of categories as default, and run the generated raster dataset recorded as Grid_Reclass. Compared with Grid_KDE with continuous floating point density values, the raster value of Grid_Reclass is a positive integer between 1 and 5 under default conditions.

[0019] S3-4. Take Grid_Reclass as the input raster, use the "Select by Attribute" tool of ArcGIS Pro, select the raster with a value greater than 1, then use the "Raster to Polygon" tool of ArcGIS Pro, use the selected raster with a value greater than 1 as the input raster, uncheck the "Simplify Polygon" option, run the generated polygon dataset and export it as a shapefile file. The polygons are the urban natural residential clusters, recorded as Shp_Cluster.

[0020] S3-5. Take Shp_Cluster as input data, use the "Feature to Point" tool of ArcGIS Pro, check the "Internal Point" option, run to generate the cluster centroid point dataset, and export the shapefile file, recorded as Shp_Center.

[0021] S3-6. Use Shp_Cluster as input data and the "Pairwise Buffer" tool of ArcGIS Pro. Set the buffer distance to W=6000 in meters (m). Uncheck "Export all features as a single feature surface". Run to generate a polygon dataset and export the shapefile file, named Shp_Buffer.

[0022] S4. The selected city (Shanghai in the embodiment) is recorded as SC, and the set of its adjacent cities (Suzhou, Jiaxing, and Nantong in the embodiment) is recorded as AC. The shapefiles storing the polygonal faces of the administrative boundaries of the above two cities are recorded as Shp_SC and Shp_AC, respectively, and the shapefiles storing the sample point data of the housing prices of the above two cities are recorded as HP_SC and HP_AC, respectively.

[0023] S5. Use Python to calculate the SSOA of the cross-border spatial contiguous area of ​​the city. The specific steps are as follows: S5-1. Call the geopandas library to read and select all points in Shp_Center that are in Shp_SC, and record this point set as SC_Center.

[0024] S5-2. Call the geopandas library, read and select all polygons containing SC_Center in Shp_Buffer, and record the polygon set as SC_Buffer.

[0025] S5-3. Call the geopandas library, read and select all polygons in Shp_Cluster that intersect with SC_Buffer, and record the polygon set as SC_Cluster.

[0026] S5-4. Call the geopandas library to read and select all polygons in SC_Cluster that are in Shp_AC. This polygon set is the cross-boundary spatial continuous area SSOA of SC.

[0027] S6. Use Python to calculate the HSSOA of the cross-border housing spillover area in the city. The specific steps are as follows: S6-1. Call the geopandas library to calculate the Euclidean distance between each polygon in the SSOA polygon set and the SC boundary, and re-sort them in ascending order.

[0028] S6-2. Initialize the housing price sample point set HP_INI = HP_SC; read the polygons in SSOA one by one in the order determined in the previous step. Repeat the following operations for each polygon P read: S6-2-1. Read and select all points in HP that are in polygon P, and merge them with HP_INI to obtain HP_Merge.

[0029] S6-2-2. Establish a hedonic price model for HP_Merge, using variables including house structure information, location characteristics, and facility distance. House structure information includes area, number of bedrooms, number of bathrooms, floors, etc. in data collection, location characteristics include road selectivity and integration calculated using spatial syntax software Depthmap, and facility distance includes the distance to hospitals, primary schools, middle schools, supermarkets and other facilities.

[0030] S6-2-3. Call the statsmodels library in Python to use the variance_inflation_factor function to calculate multicollinearity for all variables, and gradually eliminate the factors with the highest VIF (variance inflation factor) and calculate VIF again until the VIF of all variables is less than or equal to 5, and obtain the variables for constructing the final model.

[0031] S6-2-4. Call the statsmodels library in Python to use the OLS function to perform multiple linear regression, and obtain the goodness of fit R12 and R22 of HP_INI and HP_Merge respectively. If R22>R12 (that is, the goodness of fit of the model increases and the explanatory power is enhanced after adding the housing price sample data), then record the polygon P read this time and update HP_INI to HP_Merge; otherwise, do not update.

[0032] S6-3. After traversing all polygons in SSOA according to the above steps, all recorded polygons that meet the conditions are included in a set. This polygon data set is the housing spillover area HSSOA of SC.

[0033] S7. Use Python to calculate the EHSSOA of the cross-border external source impact area of ​​the city. The specific steps are as follows: S7-1. Read five types of POI point data sets located in SC, namely "hub site", "school", "tertiary hospital", "shopping center" and "scenic spot", and record them as POI_Tran, POI_Edu, POI_Hos, POI_Com and POI_Lan respectively.

[0034] S7-2, read the polygons in HSSOA one by one, and repeat the following operations every time a polygon P is read: S7-2-1. Read and select the point in Shp_Center located in polygon P, calculate the distance D1, D2...Dk from the point to the nearest k=3 points in POI_Tran / POI_Edu / POI_Hos / POI_Com / POI_Lan and use it as the new distance variable group Xadd.

[0035] S7-2-2, read and select all points in HP that are in polygon P, record them as HP_Sel; establish a characteristic price model for HP_Sel, using variables including the distance variable group calculated in the previous step, house structure information, location characteristics and facility distance. House structure information includes area, number of bedrooms, number of bathrooms, floors, etc. in data collection, location characteristics include road selectivity and integration calculated using spatial syntax software Depthmap, and facility distance includes the distance to hospitals, primary schools, middle schools, supermarkets and other facilities.

[0036] S7-2-3. Call the statsmodels library in Python to use the variance_inflation_factor function to calculate multicollinearity for all variables, and gradually eliminate the factors with the highest VIF (variance inflation factor) and calculate VIF again until the VIF of all variables is less than or equal to 5, and obtain the variables for constructing the final model.

[0037] S7-2-4. Call the statsmodels library in Python and use the OLS function to perform multivariate linear regression to obtain the fitting result of HP_Sel and determine whether the newly added distance variable Xadd is significant. If so, define the polygon P as the exogenous influence area driven by Xadd and record it, otherwise do not record it.

[0038] S7-3. After traversing all polygons in HSSOA according to the above steps, all recorded polygons that meet the conditions are included in a set. This polygon data set is the foreign aid influence area EHSSOA of SC.

[0039] S8. Call the geopandas and Matplotlib libraries in Python to perform spatial visualization of the three types of real estate market spillover areas in Shanghai: SSOA, HSSOA, and EHSSOA.

[0040] Although the embodiments of the present invention have been disclosed as above, they are not limited to the applications listed in the specification and implementation modes, and can be fully applicable to various fields suitable for the present invention. For those familiar with the art, for those of ordinary skill in the art, various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principle and spirit of the present invention. Therefore, without departing from the general concept defined by the claims and the scope of equivalents, the present invention is not limited to specific details.

Claims

1. A method for accurately measuring the spillover scope of urban real estate market based on the characteristic price model, characterized in that: The following steps are involved: S1. Collect spatial vector data of housing prices in the main built-up areas of the research city, define them as housing price sample points, and obtain a set of housing price sample points, i.e., housing price sample point set HP; the housing price sample points include: housing unit price information, housing structure information, and housing latitude and longitude information; the housing structure information includes: area, number of bedrooms, number of bathrooms, and floors; S2. Based on the housing price sample point set HP, the "Kernel Density Estimation" tool of ArcGIS Pro is used to generate a raster dataset, which is recorded as Grid_KDE; S3. Based on the raster dataset Grid_KDE, use the "Select by Attribute" tool of ArcGIS Pro to generate a set of polygon clusters, which is recorded as the urban natural residential area cluster set G; S4. Set the spatial proximity distance threshold W, and traverse each polygon in the urban natural residential cluster set G according to the following rules: for the selected polygon i, calculate its boundary distance to other polygons in G and obtain the minimum value, which is recorded as Di. If Di≤W, mark the polygon; the boundary distance is the shortest distance between the boundaries of two polygons; after the traversal is completed, all marked polygons are combined into a new set, which is the theoretical urban cross-border spatial continuous area, recorded as SSOA; S5. Select the initial city A, whose housing price sample point set is HP A According to HP A Calculate the hedonic price model M A , and obtain the initial goodness of fit R 2 A In the SSOA area of ​​the city, the housing price sample points HP corresponding to each polygon cluster i in the urban natural residential area cluster set G are arranged according to the principle of centroid distance from near to far. i Add to HP one by one A Use M again A Calculate and exclude polygonal clusters whose significant factor Beta is reversed after inclusion, and retain the goodness of fit R after inclusion 2 A The rising polygon clusters, until all SSOA regions are traversed to reach R 2 A The optimal, remaining polygonal clusters are the homogeneous housing economic spillover areas that can improve the explanatory power of the characteristic price model, denoted as HSSOA; S6. First, calculate the Euclidean distance X between each polygon cluster in HSSOA and the five important geographical features of the initial city A: medical, transportation, education, commerce, and landscape. H , X T , X E , X S , X L , and then add these five Euclidean distances as new dependent variables into the hedonic price model M A The EHSSOA of the region is further calculated based on the factor significance and the exogenous factors with key driving force are further extracted.

2. According to claim 1, a method for accurately measuring the spillover range of urban real estate market based on a characteristic price model is characterized in that: The specific steps of S3 are as follows: S3-1. Take Grid_KDE as the input grid, use the "Reclassify" tool of ArcGIS Pro, set the classification method to "Natural Breaks", keep the number of categories as default, and run to generate a grid dataset, recorded as Grid_Reclass. Compared with Grid_KDE with continuous floating point density values, the grid value of Grid_Reclass is a positive integer between 1 and 5 under default conditions. S3-2, using Grid_Reclass as input grid, use ArcGIS Pro's "Select by Attribute" tool to select grids with values ​​greater than 1, then use ArcGIS Pro's "Raster to Polygon" tool to select grids with values ​​greater than 1 as input grids, uncheck the "Simplify Polygon" option, run to generate polygon datasets and export them as shapefiles. The polygons are the natural residential clusters in the city, recorded as Shp_Cluster; S3-3. Take Shp_Cluster as input data, use the "Feature to Point" tool of ArcGIS Pro, check the "Internal Point" option, run to generate the cluster centroid point dataset, and export the shapefile file, recorded as Shp_Center.

3. The method for accurately measuring the spillover range of the urban real estate market based on the characteristic price model according to claim 2 is characterized in that: The specific steps of S4 are as follows: S4-1. Take Shp_Cluster as input data, use the "Pairwise Buffer" tool of ArcGIS Pro, set the buffer distance to the spatial proximity distance threshold W, uncheck "Export all features as a single feature surface", run to generate a polygon dataset and export the shapefile file, recorded as Shp_Buffer; S4-2, the selected city is recorded as SC, the set of its adjacent cities is recorded as AC, the shapefile files storing the polygonal surfaces of the administrative boundary ranges of SC and AC are recorded as Shp_SC and Shp_AC respectively, and the shapefile files storing the sample point data of the housing prices of the above two cities are recorded as HP_SC and HP_AC respectively; S4-3. Use Python to calculate the SSOA of the cross-boundary spatial contiguous area of ​​the city. The specific steps are as follows: S4-3-1, call the geopandas library, read and select all points in Shp_Center that are in Shp_SC, recorded as SC_Center; S4-3-2, call the geopandas library, read and select all polygons containing SC_Center in Shp_Buffer, recorded as SC_Buffer; S4-3-3, call the geopandas library, read and select all polygons in Shp_Cluster that intersect with SC_Buffer, recorded as SC_Cluster; S4-3-4. Call the geopandas library to read and select all polygons in SC_Cluster that are in Shp_AC, which is the cross-boundary spatial continuous area SSOA of SC.

4. The method for accurately measuring the spillover range of the urban real estate market based on the characteristic price model according to claim 3 is characterized by: In step S5, Python is used to calculate the urban cross-border housing spillover area HSSOA. The specific steps are as follows: S5-1. Call the geopandas library to calculate the Euclidean distance between each polygon in the cross-border spatial contiguous area SSOA and the SC boundary, and re-sort them in ascending order; S5-2, set the initial value HP_INI of the housing price sample point set HP to HP_SC; read the polygons P in SSOA one by one in the order determined in S5-1; each time a polygon P is read, perform the operations from S5-2-1 to S5-2-4: S5-2-1, read and select all points in HP that are in polygon P, and merge them with HP_INI to obtain HP_Merge; S5-2-2. Establish a hedonic price model for HP_Merge, using variables including house structure information, location characteristics and facility distance; the house structure information includes the area, number of bedrooms, number of bathrooms and floors in data collection; the location characteristics include road selectivity and integration calculated using the spatial syntax software Depthmap; the facility distance includes the distance to the hospital, primary school, middle school and supermarket; S5-2-3. Set a threshold for VIF, namely, VIF threshold; VIF is variance inflation factor; call the statsmodels library in Python to use the variance_inflation_factor function to calculate multicollinearity for all variables, and gradually remove the highest factor to calculate VIF again until the VIF of all variables is less than or equal to the VIF threshold, and obtain the variables for constructing the final model; S5-2-4. Call the statsmodels library in Python to use the OLS function to perform multiple linear regression and obtain the goodness of fit R1 of HP_INI and HP_Merge respectively. 2 and R2 2 ; If R2 2 >R1 2 , then record the polygon P read this time and update HP_INI to HP_Merge; otherwise, do not update; S5-3. After traversing all polygons in SSOA according to the above steps, all recorded polygons P that meet the conditions are defined as a set, that is, the housing spillover area HSSOA of SC.

5. The method for accurately measuring the spillover range of the urban real estate market based on the characteristic price model according to claim 4 is characterized by: In step S6, Python is used to calculate the EHSSOA of the cross-border external source impact area of ​​the city. The specific steps are as follows: S6-1, read the five types of POI point data sets located in SC, namely "hub site", "school", "tertiary hospital", "shopping center" and "scenic spot", and record them as POI_Tran, POI_Edu, POI_Hos, POI_Com and POI_Lan respectively; S6-2, read the polygons in HSSOA one by one, and execute the operations of S6-2-1 to S6-2-4 every time a polygon P is read: S6-2-1. Read and select the point in Shp_Center located in polygon P, calculate the distance from the point to POI_Tran, POI_Edu, POI_Hos, POI_Com, POI_Lan, select the three closest distances D1, D2, D3 and use them as the new distance variable group X add ; S6-2-2, read and select all points in HP that are in polygon P, record them as HP_Sel; establish a characteristic price model for HP_Sel, using variables including the distance variable group calculated in the previous step, house structure information, location characteristics and facility distance; S6-2-3. Call the statsmodels library in Python to use the variance_inflation_factor function to calculate multicollinearity for all variables, and gradually remove the factors with the highest VIF and calculate the VIF again until the VIF of all variables is less than or equal to the VIF threshold, and finally obtain the variables for constructing the model; S6-2-4. Call the statsmodels library in Python to use the OLS function to perform multiple linear regression, obtain the fitting result of HP_Sel, and determine the newly added distance variable X add Is it significant? If so, define the polygon P as affected by X. add The external source of influence area driven is recorded, otherwise it is not recorded; S6-3. After traversing all polygons in HSSOA according to the above steps, all recorded polygons that meet the conditions are included in a set. This polygon data set is the foreign aid influence area EHSSOA of SC.

6. The method for accurately measuring the spillover range of the urban real estate market based on the characteristic price model according to claim 5 is characterized by: The spatial proximity distance threshold W of step S4 is 6000; The VIF threshold in step S5-2-3 and step S6-2-3 is 5; In step S6-2-4, the distance variable X is determined. add The condition for significance is: head-to-tail split ratio p<0.05.

Images