مدلی ترکیبی برای تخمین قیمت مسکن: مطالعه موردی شهر تهران

حسینی رامندی, شیوا; کاشانی, حامد

doi:10.22065/jsce.2022.294445.2496

مدلی ترکیبی برای تخمین قیمت مسکن: مطالعه موردی شهر تهران

نوع مقاله : علمی - پژوهشی

نویسندگان

¹ دانشجوی دکتری، دانشگاه صنعتی شریف، تهران، ایران

² استادیار، دانشگاه صنعتی شریف، تهران، ایران

10.22065/jsce.2022.294445.2496

چکیده

هدف این مطالعه ایجاد مدل ترکیبی برای تخمین قیمت مسکن در شهر تهران است. تخمین قیمت مسکن از جنبه‌های متعدد مهم است. قیمت تخمینی مسکن بر نحوه‌ی برنامه‌ریزی، طراحی و ساخت مسکن تأثیر می‌گذارد. با آگاهی از قیمت تخمینی مسکن، سرمایه‌گذاران و سازندگان مسکن می‌توانند میزان سود یا ضرر خود را تخمین بزنند و با ریسک کمتری سرمایه گذاری کنند. کاهش ریسک سرمایه‌گذاری در مسکن باعث ایجاد فرصت‌های شغلی بیشتر، رشد اقتصادی و کاهش معضل‌های اقتصادی و اجتماعی ناشی از کمبود مسکن با کیفیت و قیمت مناسب می‌شود. عوامل متعددی مانند قیمت دوره‌ی قبل مسکن، تغییرات جمعیت، هزینه‌ی ساخت مسکن و تغییرات فصل تأثیر معناداری بر قیمت مسکن دارند. در این پژوهش، ابتدا عوامل کلان و اقتصاد خرد که بر قیمت مسکن تأثیر می‌گذارند بررسی و داده‌های مربوطه جمع‌آوری شد. در مرحله بعدی، داده‌ها غربال‌گری و آماده‌سازی شد. سپس، از داده‌ها برای توسعه مدل‌های رگرسیون و مدل‌های سری زمانی شامل میانگین متحرک خودهمبسته‌ی یکپارچه (آریما) و اتورگرسیون برداری استفاده شد. با استفاده از مدل رگرسیون می‌توان قیمت مسکن در سال‌های مختلف را بر اساس مجموعه‌ای از متغیرهای مستقل محاسبه کرد. مدل آریما خروجی مدل رگرسیون را دریافت می‎کند و قیمت مسکن را در سال‌های بعد تخمین می‌زند. از طرف دیگر، مدل اتورگرسیون برداری نیز می‎تواند به صورت مستقل برای تخمین قیمت مسکن استفاده شود. برای مقایسه عملکرد مدل‌ها، میزان خطای آن‌ها به دو روش درصد میانگین مطلق خطا و خطای استاندارد نسبی اندازه‌گیری شد. براساس این اندازه گیری، میزان خطای مدل اتورگرسیون برداری از ترکیب مدل رگرسیون و آریما کمتر است زیرا در مدل اتورگرسیون برداری، تأثیر متغیرهای مستقل به‌طور مستقیم در مدل اعمال شده است. مدل‌های ساخته‌شده در این پژوهش به سیاست‌گذاران، سرمایه‌گذاران، توسعه‌دهندگان، خریداران مسکن و مؤسسات مالی کمک می‌کند تا قیمت‌ مسکن را تخمین بزنند و آگاهانه در خصوص سرمایه‎گذاری در مسکن تصمیم بگیرند.

کلیدواژه‌ها

موضوعات

مدیریت پروژه و ساخت

عنوان مقاله [English]

Multiple Model for Estimating House Prices: Case Study Tehran City

نویسندگان [English]

Shiva Hoseini Ramandi ¹
Hamed Kashani ²

¹ PhD Candidate, Civil Engineering, Sharif University of Technology, Tehran, Iran

² Assistant Professor, Civil Engineering, Sharif University of Technology, Tehran, Iran

چکیده [English]

This study explores multiple mathematical models for estimating house prices in Tehran, Iran. Estimation of house prices is vital from various perspectives. Estimated prices can be used in the planning, design, and construction of residential building projects. Based on the estimated future prices of residential units, investors and constructors can conduct an investment analysis and mitigate the risk of the financial failure of residential projects. Reducing the investment risk of residential projects can lead to jobs creation and economic growth. Several factors such as previous house prices, population changes, house construction costs, and seasonal effects can significantly affect house prices. In this research, first, the macro and microeconomic factors that affect house prices were reviewed, and relevant data were collected. The next step was data preparation and pre-processing. Then, the data was used to train several models using regression and time series analysis models, namely Autoregressive Integrated Moving Average and Vector Autoregression. The current house prices can be estimated using the regression model based on a set of independent variables such as the age of the building. The ARIMA model receives the output of the regression model and estimate house price in the subsequent year. Alternatively, the Vector Autoregression model can be used independently to estimate future prices. To compare the performance of the models, their error was measured by two methods: Mean Absolute Percentage Error and Relative Standard Error. The error of the Autoregression model is less than the combination of regression and ARIMA models because in the Autoregression model, the effect of independent variables is directly applied in the model. The models developed in this research can help decision-makers, investors, developers, homebuyers, and financial institutes obtain appropriately estimated prices and make informed decisions in construction projects.

کلیدواژه‌ها [English]

Estimation
House Price
Regression
ARIMA
Vector Autoregression

مراجع

[1] Al-Zwainy, F. M. S. and Hadhal, N. T. (2015). Investigation and Evaluation of the Cost Estimation Methods of Iraqi Communication Projects. Int. J. Eng. Manag. Res., 5 (6), 41–48.

[2] Miles, W. (2008). Boom–Bust Cycles and the Forecasting Performance of Linear and Non-Linear Models of House Prices. Real Estate Financ. Econ. J..

[3] Lavasani, K. S. and AbbasiNejad, H. (2015). Forecasting the Hosing Booms or Busts Using Wavelet Decomposition and Artificial Neural Networks.

[4] Mousavi, M. and Doroodian, H. (2015). Analyzing the Determinants of Housing Prices in Tehran City. Econ. Model., 9, 103–127.

[5] Assaf, S. Bubshait, A. and Al-Muwasheer, F. (2010). Factors Affecting Affordable Housing Cost in Saudi Arabia. Int. J. Hous. Mark. Anal., 3, 290–307, doi: 10.1108/17538271011080628.

[6] Asgari, H. and Almasi, I. (2011). Factors Affecting the Price of Housing in Urban Areas Using Panel Data (1370-1385). Econ. Res. Rev., 2 (41), 201–224.

[7] Dehkordi, P. K. (2021). Analysis of the Effect of Currency Shock, Economic Sanctions and Oil Prices on the Housing Market (Using Structural Vector-Autoregressive SVAR). Appl. Theor. Econ., 7 (7), 27–56.

[8] Lowe, D. J. Emsley, M. W. and Harding, A. (2006). Predicting Construction Cost Using Multiple Regression Techniques. J. Constr. Eng. Manag., 132 (7), 750–758.

[9] Siqueira, I. (1999). “Neural Network-based Cost Estimating. Concordia University.

[10] Kim, G. H. Yoon, J. E. An, S. H. Cho, H. H. and Kang, K. I. (2004). Neural Network Model Incorporating a Genetic Algorithm in Estimating Construction Costs. Build. Environ., 39 (11), 1333–1340.

[11] Günaydın, H. M. and Dogan, S. Z. (2007). A Neural Network Approach for Early Cost Estimation of Structural Systems of Buildings. Int. J. Proj. Manag., 22 (7), 595–602.

[12] Jadevicius, A. and Huston, S. (2015). ARIMA Modelling of Lithuanian House Price Index. Int. J. Hous. Mark. Anal., 8 (1), 135–147.

[13] Amanpour, S. Soleimanirad, I. Keshtkar, L. and Mokhtari, S. (2018). Comparing the Performance of Neural Networks and Multivariate Regression in the Estimation of Housing Prices (Case Study: Ahvaz City). Geogr. Environ. Stud..

[14] Saadatmehr, M. (2011). Estimation of Hedonic Price Function for Khoramabad Urbun Housing. Knowl. Dev., 17, 313–339.

[15] Asgari, A. and Ghaderi, J. (2002). Determination of Housing Price in Urban Areas of Iran by Using Hedonic Price Method (Hpm). J. Sustain. Growth Dev. (The Econ. Res., 1 (4), 91–108.

[16] Khalili Araghi, E. Nobahar, M. (2011). Predicting Housing Prices for the City of Tabriz: Application of the Hedonic Pricing and Artificial Neural Network Models. J. Econ. Res. Policies, 19 (60), 113–138.

[17] Ghorbani, S. M. Afgheh, S. (2017). Forecasting the House Price for Ahvaz City: The Comparison of the Hedonic and Artificial Neural Network Models. J. Urban Econ. Manag., 29–44.

[18] Amanpour, S. Soleymani Rad, E. Keshtkar, L. and Mokhtari Chelcheh, S. (2015). Using Neural Network to Estimation the Cost of Housing in Ahwaz. J. Urban Econ. Manag., 3 (9), 45–57.

[19] Fazlizadeh, H. Ravanshadnia, M. and Kashani, H. (2016). Provide a Model to Predict the Business Cycles of the Housing Investment Market with the Help of Forthcoming Economic Variables Using Artificial Neural Networks and Genetic Algorithms. in Housing Economics Quarterly, p. 11.

[20] Koohpayma, J. Argany, M. and Neisany Samani, N. (2020). Apartments Price Estimation Using Linear and Geographically Weighted Regression (Case Study: District Six of Tehran City). Geogr. Urban Plan. Res., 8 (2), 347–369.

[21] Sharifi, M. (2021). Machine Learning Models for Estimating House Prices. Sharif University of Technology.

[22] Momeni, M. (2016). Assessing the Factors Affecting Housing Prices in Iran.

[23] Salemi, A. Sadrabadi, M. E. and Kimiagari, A. M. (2017). Investigating the Factors Affecting Housing Prices in Tehran in the Last Ten Years.

[24] Reybod, A. (2020). Multiple Linear Regression.

[25] Crawford G. W. and Fratantoni, M. C. (2003). Assessing the Forecasting Performance of Regime-Switching, ARIMA and GARCH Models of House Prices. Real Estate Econ., 31 (2), 223–243.

[26] Pagourtzi, E. Assimakopoulos, V. Hatzichristos, T. and French, N. (2003). Real Estate Appraisal: A Review of Valuation Methods. J. Prop. Invest. Financ., 21 (4), 383–401.

[27] Trost S. M. and Oberlender, G. D. (2003). Predicting Accuracy of Early Cost Estimates Using Factor Analysis and multivariate Regression. J. constr. Eng. Manag., 129 (2), 198–204.

[28] Reybod, A. (2021). Hierarchical Regression in SPSS.

[29] Attala M. and Hegazy, T. (2003). Predicting Cost Deviation in Reconstruction Projects: Artificial Neural Networks Versus Regression. J. Constr. Eng. Manag., 129 (4), 405–411.

[30] McGreal, S. Adair, A. McBurney, D. and Patterson, D. (1998). Neural Networks: The Prediction of Residential Values. J. Prop. Valuat. Invest., 16 (1), 57–70.

[31] Connellan O. and James, H. (1998). Estimated Realization Price by Neural Networks: Forecasting Commercial Property Values. J. Prop. Valuat. Invest., 16 (1), 71–86.

[32] Bee-Hua, G. (2000). Evaluating the Performance of Combining Neural Networks and Genetic Algorithms to Forecast Construction Demand: The Case of the Singapore Residential Sector. Constr. Manag. Econ., 18 (2), 209–218.

[33] Nguyen N. and Cripps, A. (2001). Predicting Housing Value: A Comparison of Multiple Regression Analysis and Artificial Neural Networks. J. Real Estate Res., 22 (3), 313–336.

[34] Kalami, M. (2021). Neural Networks 0 to 100.

[35] Hesaraki, E. (2019). Stock Price Forecast with ARIMA Models in Python and R - Practical Guide.

[36] Bin, O. (2004). A Prediction Comparison of Housing Sales Prices by Parametric Versus Semi-parametric Regressions. J. Hous. Econ., 13, 68–84.

[37] Hill, R. (2013). OECD Statistics Working Papers 2011 / 01 Hedonic Price Indexes for Housing.

[38] Byme, P. (1995). Fuzzy Analysis: A Vague Way of Dealing with Uncertainty in Real Estate Analysis. J. Prop. Valuat. Invest., 13 (3), 22–41.

[39] Bagnoli, C. Smith, B. and Halbert, C. (1998). The Theory of Fuzzy Logic and its Application to Real Estate Valuation. J. Real Estate Res., 16 (2), 169–200.

[40] Kusan, H. Aytekin, O. and Ozdemir, I. (2010). The Use of Fuzzy Logic in Predicting House Selling Price. Expert Syst. Appl., 37, 1808–1813.

[41] Guan, J. Zurada, J. and Levitan, A. S. (2008). An Adaptive Neuro-fuzzy Inference System Based Approach to Real Estate Property Assessment. J. Real Estate Res., 30 (4), 395–421.

[42] Asnaashari, M. Darbandi, M. Motalam, S. and Kianfar, F. (2020). Modeling Housing Prices and Examining the Factors Affecting It Using Econometric Methods and Dynamic Systems.

[43] Statistical Center of Iran. Available at: https://www.amar.org.ir/ [accessed Jun. 01, 2020].

[44] Housing Investment Group Co. Available at: https://www.maskanco.ir/ [accessed Jun. 01, 2020].

[45] R Core Team, (2020). R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria, [Online]. Available: https://www.r-project.org/.

[46] Willmott C. J. and Matsuura, K. (2005). Advantages of the Mean Absolute Error (MAE) Over the Root Mean Square Error (RMSE) in Assessing Average Model Performance. Clim. Res., 30 (1), 79–82, doi: 10.3354/cr030079.

[47] Snee, R. D. (1977). Validation of Regression Models: Methods and Examples. Technometrics, 9 (4), 415–428.

[48] Rendón, E. Abundez, I. Arizmendi, A. and Quiroz, E. M. (2011). Internal Versus External Cluster Validation Indexes. Int. J. Comput. Commun., 5 (1).