1# RECORD 309364b6d8361dba9a9e751b55ade4c452cd784b2# TITLE: The research of influence factors that possibly lead to the change in house prices3# YEAR: 2023.0 VENUE: Theoretical and Natural Science DOI: 10.54254/2753-8818/14/202409264# AUTHORS: Yi-Xiao Huang5# PAGES: 7 CANDIDATE TABLE PAGES (1-based): [3, 6]6# ABSTRACT: With the significant increase in housing prices, more research on the factors that influence housing prices is urgently needed. Most of the previous research focuses on the number data, such as the average disposable income of the population, property management fees, and the total number of the population in specific areas. In this research, the multiple linear regression model is used to judge the relationship between 10 independent variables (Carpet area, Floor, Total floor, Transaction, Furnishing, Facing, Bathrooms, Balconies, Car parking, and Ownership) and the amount of housing prices based on 910 cases. It is concluded that the ten independent variables all have strong linear relationship with the Amount. With the factors that were studied before, like the ages of the house and the policy for buying homes, these studies indeed give some ideas for the government to come up with the approach to control the increase in housing prices.789===== PAGE 1 (score 11, numeric-paren 0, table-words 0, var-words 11) =====1011The research of influence factors that possibly lead to the 12change in house prices 13Yixiao Huang 14HuaZhong University of Science and Technology, 430074, China 15 16u202316281@hust.edu.cn 17Abstract. With the significant increase in housing prices, more research on the factors that 18influence housing prices is urgently needed. Most of the previous research focuses on the number 19data, such as the average disposable income of the population, property management fees, and 20the total number of the population in specific areas. In this research, the multiple linear regression 21model is used to judge the relationship between 10 independent variables (Carpet area, Floor, 22Total floor, Transaction, Furnishing, Facing, Bathrooms, Balconies, Car parking, and Ownership) 23and the amount of housing prices based on 910 cases. It is concluded that the ten independent 24variables all have strong linear relationship with the Amount. With the factors that were studied 25before, like the ages of the house and the policy for buying homes, these studies indeed give 26some ideas for the government to come up with the approach to control the increase in housing 27prices. 28Keywords: Housing Prices, Multiple Linear Regression Model, India, Influence Factors. 291. Introduction 30The 2020 Chinese government report emphasized safeguarding the public’s access to housing. Real 31estate is one of the motivations for developing economics due to its status as a pillar industry. Based on 32the real estate revolution background, housing prices have experienced rapid growth [1]. Between 2003 33and 2013, there was a significant rise of around 13% in housing prices in China’s first-tier cities, while 34second and third-tier cities saw growth rates of 10.5% and 7.9%. These surpassed those in the US and 35Japanese real estate bubbles [2]. Meanwhile, some foreign researchers found that there was also a 36considerable increase in the housing prices in their country [3]. Simultaneously, there are many problems 37with the increasing housing prices. For instance, most youth cannot afford to buy a house, even an 38apartment. The issue of high housing prices and low income has become a challenge in urbanization 39worldwide and has attracted close attention from society, government, and academia [4]. Therefore, this 40essay aims to gain insights into the housing market dynamics and try to aid buyers, sellers, and real 41estate professionals in making informed decisions using the dataset found in Kaggle. 42In addressing the issue of rapidly rising housing prices, it is significant to understand the real estate 43market scientifically and implement relevant policies to ensure stable development of housing prices. 44Understanding the factors that influence house prices and their relationships is essential. The reasons 45behind the rise in housing prices are intricate. When studying the factors that affect the price of 46residential properties, researchers typically consider a range of variables. These may include location 47characteristics, building attributes, and the overall residential environment. Researchers may also 48Proceedings of the 3rd International Conference on Computing Innovation and Applied Physics49DOI: 10.54254/2753-8818/14/2024092650© 2023 The Authors. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.051(https://creativecommons.org/licenses/by/4.0/).52158535455===== PAGE 3 (score 23, numeric-paren 0, table-words 5, var-words 11) =====56572.2. Variable selection 58The data used in this paper count a total of 910 cases, including the house prices from 780 thousand 59rupees to 18200 thousand rupees. The data contains 11 variables: Amount (in rupees), Carpet area, Floor, 60Total floors, Transaction, Furnishing, Facing, Bathrooms, Balconies, Car parking, and Ownership. 61Table 1 shows the interpretation of each variable. Because some variables are presented in text form, 62and it is difficult for us to study, this essay changes the text form into numbers, and the corresponding 63numbers are shown in Table 1. 64Table 1. Logogram and interpretation of the 11 factors. 65Elements 66Logogram 67Interpretation 68Carpet area 69x1 70The area that can be used. Unit: square feet 71Floor 72x2 73The floor the home is in 74Total floors 75x3 76The total floor of the building that the home is in 77Transaction 78x4 79Whether the home is “Resale” or “New property”. “Resale” takes 0, “New 80property” takes 1 81Furnishing 82x5 83The condition of furnishing. “Unfurnished” take 0; “Semi-furnished” takes 841; “Furnished” takes 2 85Facing 86x6 87“East” takes 0; “West” takes 1; “North” takes 2; 88“South” takes 3; “South - West” takes 4; “North - East” takes 5; “North - 89West” takes 6; “South - East” takes 7 90Bathrooms 91x7 92The number of bathrooms 93Balconies 94x8 95The number of balconies 96Car parking 97x9 98“1 Open” takes 0; “1 Covered” takes 1; 99“2 Open” takes 2; “2 Covered” takes 3 100Ownership 101x10 102“Freehold” takes 0; “Co-operative Society” takes 1; 103“Power Of Attorney” takes 2; “Leasehold” takes 3 104Amount 105y 106The total price of the home 107Table 2 shows some statistics about the variables, including the mean, the variance, the maximum, 108and the minimum. 109Table 2. Summary statistics of each variable. 110Elements 111Logogram 112mean 113variance 114maximum 115minimum 116Carpet area 117x1 1181098.18 1191284995.59 1201820 12178 122Floor 123x2 1248.09 12532.51 12618 1271 128Total floors 129x3 13014.69 13167.68 13238 1331 134Transaction 135x4 1360.25 1370.19 138- 139- 140Furnishing 141x5 1420.61 1430.59 144- 145- 146Facing 147x6 1480.65 1492.27 150- 151- 152Bathrooms 153x7 1542.81 1551.07 1564 1571 158Balconies 159x8 1601.44 1610.55 1623 1631 164Car parking 165x9 1661.44 1671.18 168- 169- 170Ownership 171x10 1720.32 1730.69 174- 175- 176Amount 177y 1781493.58 1791202749.26 1804250 181219 182Note: The maximum (minimum) doesn’t have meaning when it comes to transactions, furnishing, 183facing, car parking, and ownership. 184Proceedings of the 3rd International Conference on Computing Innovation and Applied Physics185DOI: 10.54254/2753-8818/14/20240926186160187188189===== PAGE 4 (score 6, numeric-paren 0, table-words 2, var-words 2) =====1901912.3. Research protocol 192This paper uses the Linear Regression model and Single Factor Correlation Analysis, the total amount 193of the home is the dependent variable(Y), and the 10 factors are the independent variables(X). Next, this 194paper uses SPSSAU to analyze the relationship between the effect of X on Y based on the results of 195scatter plots, box plots and multiple linear regression. 1962.4. Model principle 197Linear regression is a standard statistical learning method to establish a linear relationship between the 198independent and dependent variables. Here is the typical formula for linear regression: 199y = β0 + β1 x1 + β2 x2 + ⋯+ βp xp + ε200(1) 201Where β is the coefficient, ε is the residuals and the expectation for ε is 0. 202This paper uses scatter plots to directly see the relationship between two distinct variables. 2033. Results and Discussion 204Figure 1 shows that it has an increasing trend, leading us to use a linear regression model. The result is 205y = 955.730 + 0.490 ∗x1, and the R square is 0.256. Because the R square is very close to 0, the 206dataset should delete the outliers to keep the R square close to 1. 207 208Figure 1. The scatter plot of carpet area and the amount. 209As shown in figure 2, after deleting all the outliers, the data seems more linear. However, the results 210show that the regression model becomes y = −301.308 + 1.705 ∗x1, and R square becomes 0.476, 211still not very close to 1. 212 213Figure 2. Processed figure 1. 214Proceedings of the 3rd International Conference on Computing Innovation and Applied Physics215DOI: 10.54254/2753-8818/14/20240926216161217218219===== PAGE 6 (score 40, numeric-paren 12, table-words 6, var-words 10) =====220221that all the factors have relatively strong affection on the amount. Moreover, the p-values of Carpet area, 222Floor, Total floor, Transaction, Furnishing, Bathrooms, Balconies, Car parking, and Ownership are 223smaller than 0.01, which illustrates these variables have significantly strong relationship with the 224Amount. When it comes to the collinearity diagnosis, the values of tolerance of Carpet area, Floor, Total 225floor, Transaction, Furnishing, Facing, Bathrooms, Balconies, Car parking, and Ownership all are 226greater than 0.02. The outcomes show no autocorrelation in the model and no correlation between the 227sample data, indicating that the model is of good quality. Moreover, the D-W value is around 2, which 228also illustrates that the model doesn’t have autocorrelation or correlation between the sample data. The 229second column shows the coefficients of 10 variables. The coefficients of Carpet area, Floor, Total floor, 230Furnishing, Facing, Bathrooms, Balconies, Car parking, and Ownership are all bigger than 0, with the 231result of p-values, which means these variables have significant favourable influence on the amount. 232The coefficient of Transaction is less than 0, with the consequence of p-value, so this variable has a 233significant negative impact on the amount, which means that the prices of “Resale” are universally 234greater than “New property”. 235In comparison to prior research, they tend to use specific analysis of a single factor, especially the 236factors shown in number, which is limited to known possible factors for influencing the housing prices, 237such as Facing, Transaction, Ownership, Furnishing, and Car parking, the data initially shown in the text 238form. In contrast, this study covers a wide range of factors, making it more comprehensive and effective 239in avoiding errors caused by not controlling for a single variable. It also expands ideas for future research 240on housing prices and helps sellers identify the preferences of potential buyers. 241Table 3. A slightly more complex table with a narrow caption. 242 243B 244t 245p 246collinearity diagnosis 247VIF 248tolerance 249constant 250-1347.29 251-14.413 2520.000** 253- 254- 255x1 2560.085 2573.645 2580.000** 2592.099 2600.476 261x2 26217.03 2633.483 2640.001** 2652.273 2660.44 267x3 26829.62 2698.702 2700.000** 2712.304 2720.434 273x4 274-314.184 275-6.454 2760.000** 2771.342 2780.745 279x5 280353.179 28112.373 2820.000** 2831.42 2840.704 285x6 28628.751 2872.18 2880.029* 2891.169 2900.855 291x7 292431.747 29311.769 2940.000** 2954.236 2960.236 297x8 298134.352 2994.934 3000.000** 3011.226 3020.816 303x9 304403.357 30512.619 3060.000** 3073.57 3080.28 309x10 31073.178 3113.032 3120.002** 3131.207 3140.829 315R2 3160.74 317adjustment R2 3180.737 319F 320F (10,889)=253.086,p=0.000 321D-W value 3221.822 323Dependent: y 324* p<0.05 ** p<0.01 3254. Conclusion 326In this study, various data were analysed to determine the factors that affect housing prices. The results 327suggest that changes in housing prices are linked to factors such as carpet area, floor, total number of 328floors, transaction, furnishing, facing direction, number of bathrooms and balconies, availability of 329Proceedings of the 3rd International Conference on Computing Innovation and Applied Physics330DOI: 10.54254/2753-8818/14/20240926331163332333334===== PAGE 7 (score 5, numeric-paren 0, table-words 0, var-words 5) =====335336parking space, and ownership. Most of the factors studied in this paper have not received much attention 337in previous studies. 338It is true that the model may have errors due to the limited amount of data. Additionally, the sample 339did not cover all conditions from different countries or discuss policy and economic backgrounds, which 340could cause possible variations and affect the accuracy of the results. The study still has some advantages. 341Firstly, the data was from India, and was just released last month, which ensures the uniqueness of our 342data. Secondly, this essay chooses the data in text form and changes them into number form, which is 343really innovative. Thirdly, this essay uses more than one method to get the outcomes. It ensures the 344feasibility of the linear regression model and eliminates errors. At last, we draw a conclusion that the 10 345variables all have relationship with the amount of the housing prices, but it is still a preliminary 346conclusion. Further investigation is needed to determine if these factors are linked to changes in housing 347prices, and these findings may guide future research. Once other factors that influence housing prices 348are found, the actions that will be taken to control the prices may come out more quickly and more 349efficiently. 350References 351[1] 352Xie R, Du Y and Zhu J M 2022 Econometric Analysis of the Factors Influencing Housing Prices 353in Anhui based on Multiple Regression. Natural Science Journal of Harbin Normal University 35438, 36-42. 355[2] 356Chen G and Lu Y X 2022 Examining the Relationship between Confucian Tradition and Real 357Estate Price Growth: The Realization of Myriad Spacious Mansions. China Journal of 358Economics 9, 236-266. 359[3] 360Ernawati M K, Hasnanywati H and Atasya O2016 Factors Influencing the Housing Price: 361Developers’ Perspective World Academy of Science Engineering and Technology. 362International Journal of Humanities and Social Sciences, 10, 1676-1682. 363[4] 364Wu B R 2009 India Adopts New Measures to Address High Housing Prices. Huadong Technology, 36564-65. 366[5] 367Qin J R, et al. 2021 Analysis of the Influencing Factors of Housing Price in Nanjing. Journal of 368Geo-information Science, 23, 882-890. 369[6] 370Liu R G, You J Q and Zhang Y 2022 Analysis of Housing Price Influencing Factors Based on 371Factor Analysis Method. Inner Mongolia Science Technology & Economy 5, 45-60. 372[7] 373Cui Z, Zhou M Q and Kong 2022 Lingzhu Research on the Heterogeneity of Factors Affecting 374Urban Housing Prices in China. Taxation and Economy Serial 245, 65-74. 375[8] 376Sun L F and Zhang S D 2014 Research on the Relationship between International House Prices, 377Economic Growth, and Loan Interest Rates Based on GMM Estimation Times. Finance 378Cumulatively 567, 40-41. 379[9] 380Ma Y and Kong J 2015 Population aging, Economic Growth, and House Price Fluctuations: 381Empirical Analysis of International Experience Based on the Model of Generational Overlap. 382China Academic Journal Electronic Publishing House 15, 237-262. 383[10] Guo Y, Yu H S and Ke J W 2022 Influencing Factors of Second-hand Housing Prices in Yantai 384City Based on Covariance Regression. Journal of Ludong University (Natural Science Edition) 38538, 232-240. 386Proceedings of the 3rd International Conference on Computing Innovation and Applied Physics387DOI: 10.54254/2753-8818/14/20240926388164389