# RECORD 8506f16b0710a57ed835f9583338b207fd204258 # TITLE: Analysis of Tianjin real estate market: multiple linear regression prediction method # YEAR: 2024.0 VENUE: Transactions on Economics, Business and Management Research DOI: 10.62051/a4tma782 # AUTHORS: J. Yi # PAGES: 5 CANDIDATE TABLE PAGES (1-based): [4] # ABSTRACT: In this paper, the price prediction of Tianjin real estate market is studied, and the key factors affecting housing prices are systematically analyzed by using the method of multiple linear regression prediction. The research data comes from the real estate transaction data of Tianjin from 2018 to 2022, covering independent variable information such as housing area, housing age, geographical location and supporting facilities. By constructing multiple linear regression model, it is found that housing area and supporting facilities have a significant positive impact on housing prices, while housing age and distance from the city center have a significant negative impact on housing prices. Specifically, for every 1 square meter increase in housing area, house prices are expected to rise by 450 yuan; Every year the age of the house increases, the house price is expected to drop by 200 yuan; Every kilometer away from the city center, house prices are expected to drop in 800 yuan; Every time the score of supporting facilities is increased by 1 point, the house price is expected to increase by 1200 yuan. The goodness-of-fit test of the model shows that the R value is 0.85, which shows that the model has strong explanatory power. In addition, through the correlation analysis of independent variables, it is found that there is a positive correlation between housing area and supporting facilities, but the correlation with other variables is weak. This study not only enriches the theoretical research of the real estate market, but also provides valuable decision-making reference for the government, enterprises and consumers, and helps the healthy and stable development of the real estate market in Tianjin. ===== PAGE 1 (score 3, numeric-paren 0, table-words 0, var-words 3) ===== Transactions on Economics, Business and Management Research ISSN: 2960-1789, eISSN: 2960-2254 | Volume 12 EDI 2024 Content from this work may be used under the terms of CC BY-NC 4.0 licence (https://creativecommons.org/licenses/by-nc/4.0/). Published by Warwick Evans Publishing. WEP Warwick Evans Publishing Analysis of Tianjin real estate market: multiple linear regression prediction method Juntai Yi Changjun High School International Department, Changsha, 410002, China Abstract. In this paper, the price prediction of Tianjin real estate market is studied, and the key factors affecting housing prices are systematically analyzed by using the method of multiple linear regression prediction. The research data comes from the real estate transaction data of Tianjin from 2018 to 2022, covering independent variable information such as housing area, housing age, geographical location and supporting facilities. By constructing multiple linear regression model, it is found that housing area and supporting facilities have a significant positive impact on housing prices, while housing age and distance from the city center have a significant negative impact on housing prices. Specifically, for every 1 square meter increase in housing area, house prices are expected to rise by 450 yuan; Every year the age of the house increases, the house price is expected to drop by 200 yuan; Every kilometer away from the city center, house prices are expected to drop in 800 yuan; Every time the score of supporting facilities is increased by 1 point, the house price is expected to increase by 1200 yuan. The goodness-of-fit test of the model shows that the R value is 0.85, which shows that the model has strong explanatory power. In addition, through the correlation analysis of independent variables, it is found that there is a positive correlation between housing area and supporting facilities, but the correlation with other variables is weak. This study not only enriches the theoretical research of the real estate market, but also provides valuable decision-making reference for the government, enterprises and consumers, and helps the healthy and stable development of the real estate market in Tianjin. Key words: Tianjin; real estate market; multiple linear regression. 1. Introduction As an important municipality directly under the central government and economic center of China, the development of Tianjin's real estate market has an important impact on the regional and even the national economy. In recent years, with the acceleration of urbanization and the improvement of residents' living standards, the real estate market in Tianjin has shown a vigorous development trend, attracting a large number of investors and residents' attention. However, the volatility and uncertainty of the real estate market also bring a lot of risks to relevant stakeholders [1]. Therefore, in-depth analysis and accurate prediction of Tianjin real estate market will not only help the government and enterprises to make scientific and reasonable decisions, but also provide consumers with more transparent market information. The purpose of this study is to make a systematic analysis of Tianjin real estate market by using the method of multiple linear regression prediction. As a statistical method, multiple linear regression can explore the relationship between multiple independent variables and dependent variables, and help us understand the influencing factors of house prices more comprehensively. This study not only has theoretical significance, but also has practical application value. Through scientific forecasting methods, we can better grasp the market dynamics and contribute to the healthy and stable development of Tianjin real estate market. ===== PAGE 4 (score 23, numeric-paren 0, table-words 8, var-words 4) ===== 36 Table 1 Prediction results of multivariate linear regression model variable B SE T value P value Intercept 50000 15000 3.33 <0.01 Housing area (X1) (per square meter) 450 35 12.86 <0.001 Room age (X2) (per year) -200 60 -3.33 <0.01 Geographical location (X3) (per kilometer) -800 120 -6.67 <0.001 Supporting facilities (X4) (each score) 1200 150 8.00 <0.001 4.2. Independent variable correlation analysis The correlation between independent variables is also considered in the process of model construction. By calculating the correlation coefficient matrix, it is found that there is a certain degree of positive correlation between housing area and supporting facilities, which may be because larger houses are often equipped with more perfect facilities. The correlation between other independent variables is weak, which will not cause serious multicollinearity problems to the model. See Table 2. Table 2 Correlation coefficient matrix variable Housing area (X1) Room age (X2) location (X3) Supporting facilities (X4) Housing area (X1) 1.00 -0.10 -0.20 0.40 Room age (X2) -0.10 1.00 0.05 -0.05 Geographical location (X3) -0.20 0.05 1.00 -0.15 Supporting facilities (X4) 0.40 -0.05 -0.15 1.00 There is a certain degree of positive correlation between the housing area (X1) and the supporting facilities (X4), and the correlation coefficient is 0.40. This means that larger houses are often equipped with better facilities. This positive correlation is reasonable in the real estate market, because large houses usually have more space and resources to provide comprehensive supporting facilities, thus improving the convenience and comfort of living. There is a certain negative correlation between the housing area (X1) and the geographical location (X3), and the correlation coefficient is -0.20. This may indicate that the housing area in Tianjin, which is closer to the city center, may be relatively small due to limited land resources. On the contrary, far away from the city center, land resources are more abundant, so the housing area may be larger. However, this negative correlation is relatively weak, indicating that the relationship between area and geographical location is not absolute. Room age (X2), its correlation with other variables is very weak. Especially, the correlation coefficients with the building area (X1) and supporting facilities (X4) are only -0.10 and -0.05, which are almost negligible. This means that the house age may be a relatively independent variable, which is less affected by other factors. In fact, the age of the house mainly reflects the old and new degree of the house, which is not directly related to the size of the house and the perfection of supporting facilities. Finally, there is a certain degree of negative correlation between geographical location (X3) and supporting facilities (X4), and the correlation coefficient is -0.15. This may indicate that in Tianjin, in areas far from the city center, supporting facilities may be relatively few or not perfect. However, ===== PAGE 5 (score 5, numeric-paren 1, table-words 0, var-words 4) ===== 37 this negative correlation is relatively weak, indicating that the relationship between geographical location and supporting facilities is not static. The multiple linear regression model in this study is suitable for forecasting the housing price in Tianjin under similar market conditions. However, the model also has some limitations. First of all, the model assumes that there is a linear relationship between independent variables and dependent variables, which may not be fully established in practical application. Secondly, the model does not consider all the factors that may affect housing prices, such as market supply and demand, policy changes and so on. Therefore, it is necessary to combine other information to make a comprehensive judgment in practical application. To sum up, this study successfully predicted the housing price in Tianjin through multiple linear regression model, and analyzed the influence degree of their respective variables on housing prices and their correlation. Although the model has certain forecasting ability and application scope, it still needs to pay attention to its limitations and consider them in practical application. 5. Conclusion In this study, the real estate market in Tianjin is systematically analyzed by using the method of multiple linear regression prediction. The results show that housing area, housing age, geographical location and supporting facilities are the key factors affecting housing prices in Tianjin. Among them, housing area and supporting facilities have a significant positive impact on housing prices, while housing age and distance from the city center have a significant negative impact on housing prices. In addition, the correlation analysis between independent variables shows that there is a certain degree of positive correlation between housing area and supporting facilities, while there is a certain degree of negative correlation between housing area and geographical location. These findings provide strong support for the healthy and stable development of Tianjin real estate market. Although the multiple linear regression model has certain forecasting ability and applicable scope, it still needs to pay attention to its limitations and consider them in practical application. First of all, the model assumes that there is a linear relationship between independent variables and dependent variables, which may not be fully established in practical application. Therefore, when using this model to predict, it is necessary to combine other information to make a comprehensive judgment. Secondly, the model does not consider all the factors that may affect housing prices, such as market supply and demand, policy changes and so on. Therefore, in practical application, it is necessary to further collect and sort out relevant data in order to predict house prices more accurately. Finally, in order to improve the accuracy of forecasting, other more advanced forecasting methods, such as machine learning and artificial intelligence, are considered, so as to provide more comprehensive and accurate decision- making basis for the development of Tianjin real estate market. References [1] Su Zhi. (2016). Short-term correlation between public expectations and the real estate market - An empirical study based on microblog information. Economic and Management Research, 37(3), 8. [2] Wu Tingting, Hu Wenxiu, & Zhao Fan. (2018). Dynamic early warning research on bubble economic crises – Based on international experience data from the real estate market. Forecasting, 37(3), 7. [3] Cui Mingming, Liu Xiaoting, Li Xiuting, & Dong Jichang. (2020). Integrated forecasting of the real estate market driven by data characteristics. Management Review, 32(7), 13. [4] Wang Shaofen, & Cai Chengbin. (2020). Grey relational analysis of the impact of population structure changes on real estate market demand. Practice and Cognition of Mathematics, 50(9), 5. [5] Liu Xiaojun, Hu Shengkai, & Chi Yihan. (2021). Real estate price prediction research based on var—gm(1.1)—svr model. Practice and Cognition of Mathematics, 51(1), 12. [6] Shao Weishuang, Li Xiaohong, Zhang Tianshu, & Wang Yan. (2020). Application research of data mining in real estate price prediction. Practice and Cognition of Mathematics, 50(5), 6.