1# RECORD 4e466a17fb2cf5a4adc290cec91b0f1aaedbabd12# TITLE: Research on the Influencing Factors of Rental House Prices3# YEAR: 2024.0 VENUE: Transactions on Economics, Business and Management Research DOI: 10.62051/cs0pd7284# AUTHORS: Xiaoyan Mao5# PAGES: 6 CANDIDATE TABLE PAGES (1-based): [3, 4]6# ABSTRACT: Since the 1990s, China's real estate market has gradually recovered, and has developed into a large and mature trading market today, as the housing price increase, the demand of housing rental also increases a lot. Based on this this article aims to identify the factors that have an impact on rental price. The method of multi-linear regression is used to analyze the factors with over 10,000 samples in Guangzhou from May 2020 to June 2020 from the website of LianJia, one of the biggest real estate agencies in China. The dataset contains 10 variables in total, this research uses VIF, Pearson correlation coefficient to narrow the variables down to District, Rent Type, Area, Elevator Flag and Buld Floor these five variables and considered that they have a significant linear relationship with rental price. By focusing on these variables this study provides a view of the relationship of demand and supply of China’s renting market.789===== PAGE 1 (score 18, numeric-paren 0, table-words 2, var-words 14) =====1011 12Transactions on Economics, Business and Management Research 13ISSN: 2960-1789, eISSN: 2960-2254 | Volume 10 14MEEA 2024 15 16Content from this work may be used under the terms of CC BY-NC 4.0 licence (https://creativecommons.org/licenses/by-nc/4.0/). 17Published by Warwick Evans Publishing. 18WEP19Warwick20Evans21Publishing22Research on the Influencing Factors of Rental House Prices 23Xiaoyang Mao 24Chemistry Department, Denison University, Granville, 43023, The US 25mao_x1@denison.edu 26Abstract. Since the 1990s, China's real estate market has gradually recovered, and has developed 27into a large and mature trading market today, as the housing price increase, the demand of housing 28rental also increases a lot. Based on this this article aims to identify the factors that have an impact 29on rental price. The method of multi-linear regression is used to analyze the factors with over 10,000 30samples in Guangzhou from May 2020 to June 2020 from the website of LianJia, one of the biggest 31real estate agencies in China. The dataset contains 10 variables in total, this research uses VIF, 32Pearson correlation coefficient to narrow the variables down to District, Rent Type, Area, Elevator 33Flag and Buld Floor these five variables and considered that they have a significant linear 34relationship with rental price. By focusing on these variables this study provides a view of the 35relationship of demand and supply of China’s renting market. 36Keywords: Rental price; Linear regression; Pearson correlation coefficient. 371. Introduction 38Since the housing reform in the 1990s, China's housing rental market has also begun to recover under 39the strong market, demographic fundamentals and strong support from the government [1]. In recent 40years, under the influence of many factors, the rental population in the country shows an increasing 41trend year by year [2]. With the expansion of the rental market and the continuous growth of the 42number of tenants, the study of the house rental price has become increasingly important. This will 43not only help strengthen the supervision of the rental market, but also provide tenants with better 44pricing references. 45In recent years, with the promotion of national policies and the continuous improvement of the market, 46the housing rental market in Guangzhou has gradually matured, but it also faces many challenges [3]. 47On the one hand, the fluctuation of housing rental prices is affected by many factors, such as 48geographical location, housing quality, supporting facilities, etc. [4]. On the other hand, the demand 49for housing is increasingly diversified, and there are differences in the expectations and affordability 50of housing rental prices. Therefore, this paper aims to provide a reference for the government to 51formulate relevant policies, for developers to optimize product positioning, and for residents to 52rationally choose housing by systematically analyzing the influencing factors of housing rent prices 53in Guangzhou. 54In related research fields, many scholars have conducted in-depth discussions on the influencing 55factors of housing rental prices. From the perspective of building characteristics, factors such as house 56size, orientation, floor, house age, elevator, and property management will all have an impact 57on rental price [5]. Another key factors that significantly impact rental prices in Guangzhou is 58transportation convenience. The city's robust public transportation system, including its extensive 59subway network and bus routes, has a profound effect on the accessibility and attractiveness of 60different neighborhoods. Properties located near major transportation hubs tend to command higher 61rents due to their superior connectivity and convenience. This is particularly true for commuters and 62young professionals who value efficiency and time savings [6]. In addition, environmental factors 63also play a crucial role in determining rental prices. The accessibility of parks and green spaces, as 64well as the level of industrial pollution, can significantly influence the desirability and pricing of 65housing units. Residents are increasingly aware of the importance of living in a healthy and 66sustainable environment, and this is reflected in the rental market. Properties situated in areas with 676869===== PAGE 3 (score 35, numeric-paren 3, table-words 11, var-words 3) =====7071 72148 732.3. Method Introduction 74This paper uses Pearson correlation coefficient to test the correlation between the price and other 75variables. This test is mainly aim for the correlation between variables and select the variables can be 76used in the further multi-linear regression model. The multi-linear regression model is also used in 77this paper. It is a linear regression model with more than one variable. It is used in explain the linear 78relationship between one variable and multiple related variables to make the model closer to the 79reality. 80 81𝐸(𝑌) = 𝛽0 + 𝛽1𝑥1 + 𝛽2𝑥2 + ⋯+ 𝛽𝑝𝑥𝑝+ 𝑒 (1) 82 83In the above formula: 𝛽0 is a constant term, and 𝑒 is a residual term. 843. Results and Discussion 853.1. Multi-linear Regression 86This analysis shows that there are many factors can impact on rental price. As Table 2 shows: 87Table 2. Multi-linear Regression analysis on six variables 88 89Nonnormalized 90Coefficient 91Normalized 92Coefficient 93t 94p 95Collinearity 96B 97S.E. 98Beta 99VIF Tolerance 100Constant 101-360.505 102377.670 103- 104-0.955 0.340 105- 106- 107District (x1) 10884.865 10910.472 1100.057 1118.104 0.000**1.0161120.984 113RentType (x2) 114-70.768 115184.130 116-0.003 117-0.384 0.701 1.0251180.975 119Area (x3) 12041.938 1210.445 1220.664 12394.1420.000**1.0301240.970 125BuldFloor (x4) 12635.954 1273.518 1280.089 12910.2210.000**1.5591300.642 131ElevatorFlag (x5) 132129.228 13381.439 1340.014 1351.587 0.113 1.5211360.658 137CommunityID 138(x6) 1390.009 1400.027 1410.003 1420.348 0.728 1.0991430.910 144 145Based on the analysis above the relevant multi-linear regression equation is available: 146 147𝑃𝑟𝑖𝑐𝑒= −360505 + 84.865𝑥1 −70.768𝑥2 + 41.938𝑥3 + 129.228𝑥4 + 35.954𝑥5 + 0.009𝑥6 (2) 148 149It is showed in the regression analysis that the coefficient value of District is 84.865, and t is equal to 1508.104, p is 0.000, smaller than 0.01, which indicate that District has a significant positive impact on 151Price. The coefficient value of Rent Type is -70.768, and t is -0.384, p is 0.701, larger than 0.05, it 152shows that Rent Type has no significant effect on Price. The coefficient value of Area is 41.938, with 153t equals 94.142 and p equals 0.000, which is samller than0.01, indicating that Area has a significant 154positive effect on Price. The coefficient value of Buld Floor is 35.954, t is 10.221, and p is 0.000, 155smaller than 0.01, which means that Buld Floor has a significant positive effect on Price. The 156coefficient value of Elevator Flag is 129.228 it has a t equal 1.587 and p equals 0.113, which is larger 157than 0.05, which indicates that Elevator Flag has no significant effect on Price. The coefficient value 158of Community ID is 0.009 with a t value of 0.348 and p value of 0.728, larger than 0.05, this shows 159that Community ID doesn’t have a significant influence on Price. To sum up, District, Area and Buld 160Floor have a significant positive impact on Price, while Rent Type, Elevator Flag and Community ID 161have no significant impact on Price 162163164===== PAGE 4 (score 40, numeric-paren 8, table-words 12, var-words 2) =====165166 167149 1683.2. Data Relevance 169The regression analysis shows that only district, area and buld floor have a significant positive impact 170on the rent price. Pearson Correlation Coefficient is introduced to double check the correlation 171between rent price and the independent variables. From Table 3, it demonstrates the Pearson 172correlation coefficient of the independent variables with respect to price. Given the complexity of 173geometry and house floor in the dataset, they are not incorporated in the analysis. Furthermore, the 174unit price equates to price per area and is solely reliant on price and area, excluding it from the test. 175The table above allows for determining the correlation between price and the independent variables. 176The correlation between price and region is 0.070, significant at the 0.01 level, indicating a significant 177positive correlation between the two. The correlation between price and lease type is 0.086, also 178significant at the 0.01 level, suggesting a significant positive correlation. The correlation between 179price and elevator sign is 0.107, significant at the 0.01 level, signifying a significant positive 180correlation. The correlation between price and area is 0.671, significant at the 0.01 level, indicating a 181significant positive correlation. The correlation between price and building floor is 0.147, significant 182at the 0.01 level, suggesting a significant positive correlation. The correlation between price and 183community ID is -0.001, near 0, with a P-value of 0.877, greater than 0.05, suggesting no correlation 184between price and community ID. In conclusion, district, rent type, elevator flag, area, and building 185floor all exhibit a significant correlation with the rent price. However, the community ID doesn’t 186show any significant correlation to the rent price, so it can be ignored in further analysis. 187Table 3. Pearson correlation coefficient analysis on variables 188 189Price 190District (x1) 1910.070** 192RentType (x2) 1930.086** 194Area (x3) 1950.671** 196BuldFloor (x4) 1970.147** 198ElevatorFlag (x5) 1990.107** 200CommunityID (x6) 201-0.001 2023.3. Re-build Multi-linear Regression Model 203According to the Pearson correlation coefficient a new multi-linear regression model was built. 204District, Rent Type, Area, Elevator Flag and Buld Floor are the remaining independent variables 205(Table 4). 206Table 4. Multi-linear Regression Analysis five remaining variables 207 208Nonnormalized Coefficient Normalized Coefficient 209t 210p 211Collinearity 212B 213S.E. 214Beta 215VIF Tolerance 216Constant 217-346.032 218375.352 219- 220-0.922 0.357 221- 222- 223District (x1) 22484.809 22510.471 2260.057 2278.100 0.000** 1.016 0.984 228RentType (x2) 229-68.823 230184.037 231-0.003 232-0.374 0.708 1.024 0.976 233Area (x3) 23441.947 2350.445 2360.664 23794.328 0.000** 1.027 0.974 238ElevatorFlag 239(x4) 240125.881 24180.864 2420.013 2431.557 0.120 1.500 0.667 244BuldFloor (x5) 24535.757 2463.471 2470.088 24810.300 0.000** 1.518 0.659 249 250Based on the new regression model, new equation can be obtained: 251 252𝑃𝑟𝑖𝑐𝑒= −346.032 + 84.809𝑥1 −68.823𝑥2 + 41.947𝑥3 + 125.881𝑥4 + 35.757𝑥5 (3) 253 254255256===== PAGE 5 (score 17, numeric-paren 0, table-words 6, var-words 4) =====257258 259150 260In the table 4, no variable has a VIF greater than 5, which indicate that there is no covariance problems 261occur. 262Table 5. Multi-linear Regression Model Summary 263R 264R2 265Adjusted R2 266RMSE 267DW 268AIC 269BIC 2700.681 2710.464 2720.463 2733252.976 2741.570 275211469.338 276211513.238 277 278According to the data in the above table 5, the square value of model R is 0.464, which indicates that 279factors such as District, Rent Type, Area, Elevator Flag and Buld Floor can explain 46.4% of the 280change in Price. Moreover, during the F-test of the model, the result shows that the model passes the 281F-test (F=1920.929, p=0.000<0.05), which indicates that the construction of the model has practical 282significance. 2834. Conclusion 284The study selected 11128 samples from LianJia from May to June 2020, which contains 10 variables. 285This paper uses multi-linear regression and Pearson correlation coefficient combine with logical 286reasoning to reduce the variables to 5 and conclude that District, Rent Type, Area, Elevator Flag and 287Buld Floor have a significant impact on the rent price. 288It cannot be denied that due to the change of the policy, supply and demand relationship and 289customer’s expectation this model cannot perfectly fit in the real market situation. Also, the data for 290the district is not evenly distributed, which may also make the model has a district preference. 291However, this study still has a lot of value in some aspects. First of all, the method used in this study 292is innovative, the use of multi-linear regression is to find out the relationship between variables and 293the rental price. Compare to the single factor analysis method it makes the analysis more 294comprehensive and closer to the reality. It provides an insight into the proportionate weight of 295location and the building itself in rental pricing. On one hand, the analysis of this model reflects 296customer’s demand for rent from the side, which provides the direction for the housing agent when 297choosing the high-quality housing. This can improve the relationship between supply and demand of 298rent market, making the housing agent choose more targeted housing and lower down the cost of 299choosing low demand housing. On the other hand, this study shows the relationship between variables 300and rental price, which also gives the customer a measuring standard when renting house. This can 301help the customer balance the housing demand and cost, lower down the waste of money on unwanted 302requirements. In general, this study gives the supplier and customer in the housing rent market a 303guidance on the demand and supply, which can improve the market efficiency and reduce the cost of 304unwanted demand. 305References 306[1] W. Y. Zhang, People Living in rental residential housing in China 2017-2023, Statista, (2022). 307[2] Y. Wang, et al. Spatial Differentiation and Influencing Factors Dataset of Housing Rents in Guangdong-Hongkong-308Macao Greater Bay Area, Journal of Global Change Data & Discovery, 6 (1) (2022) 37-44. 309[3] Y. Wang, et al. The core influencing factors of housing rent difference in Guangzhou’s urban district, Acta 310Geographica Sinica, 76(8) (2021). 311[4] L. Q. Gu, On the perfection of China’s Long-term Rental System, Frontiers in Business, Economics and 312Management, (2022). 313[5] T. Hu, The rise of China’s rental housing market, Jones Lang LaSalle, (2018). 314[6] D. G. Wang, S. M. Li, Socio-economic differentials and stated housing preferences in Guangzhou, China, Habitat 315International, (2006). 316[7] Y. X. Xu, S. J. He, J. X. Qian, An investigation on the emerging housing rental market in a studentified village in 317Guangzhou: a new institutional economics perspective. University at Albany, (2020). 318