Iranian Journal of Economic Research

An Analysis of Speculators’ Behavior in the Iranian Housing Market Through Agent-Based Modeling and Genetic Algorithm

Document Type : Research Paper

Authors

Neda Alamolhoda 1
Marjan Damankeshideh 2
Meysam Amiri 3
Amir Reza Keyghobadi 4

1 Ph.D. Candidate in Economic Sciences, Faculty of Economics and Accounting, Central Tehran Branch, Islamic Azad University, Tehran, Iran

2 Associate Professor, Department of Economics, Central Tehran Branch, Islamic Azad University, Tehran, Iran

3 Department of Finance and Banking, Allameh Tabataba’i University, Tehran, Iran

4 Associate Professor, Department of Industrial Management, Central Tehran Branch, Islamic Azad University, Tehran, Iran

https://doi.org/10.22054/ijer.2026.86895.1379
Abstract
Housing is both a basic necessity and a non-substitutable commodity that possesses a dual nature, functioning simultaneously as a consumption good and a durable capital asset. Relying on the agent-based modeling, the present study aimed to develop a systemic model of housing price formation that could incorporate heterogeneous agents, including consumer-buyers, investors, and real estate builders. Using quarterly data for Iran spanning the period from 1998 to 2022, the proposed model was estimated through a genetic algorithm. The results of the shock analysis indicated that speculative investor behavior played a crucial role in driving housing price dynamics and volatility. In the absence of speculative investors, housing prices would converge toward a fundamental equilibrium determined by production costs and effective demand. However, speculative behavior could generate persistent deviations from this equilibrium and even trigger explosive increases in housing prices. According to the empirical findings, the Iranian housing market exhibited a high degree of forward-looking behavior (52.16%) and bubble-like, unstable dynamics throughout the sample period. From a policy perspective, the model showed that the mortgage loan-to-price ratio was not an effective instrument for stabilizing the housing market during either boom or recession periods, a finding that is consistent with real-world market evidence. By contrast, reducing construction costs could improve market rationality, moderate the responsiveness of forward-looking investors, enhance price stability, stimulate market activity, and ultimately contribute to a more stable housing market.
Introduction
The housing market is one of the most significant economic sectors. Owing to its depth and extensive linkages with a wide range of industries, it generates substantial spillover effects throughout the broader economy. Furthermore, fluctuations in housing prices can significantly influence consumption and investment behavior, ultimately shaping overall economic performance. In addition to serving as a consumption good, housing is also regarded as a capital asset. Demand for housing driven by consumption needs generally follows a relatively stable and predictable pattern. In contrast, investment demand is more sensitive to market conditions and is therefore considerably more volatile.
These characteristics make the analysis of the housing market particularly challenging and underscore the need for more sophisticated modeling approaches. In this respect, the present study employed an agent-based modeling grounded in behavioral economics to provide a deeper understanding of housing market dynamics and improve the prediction of market behavior. Agent-based models are well suited for analyzing and simulating housing markets because they explicitly capture the heterogeneous behavior of economic agents and the complex interactions among them. Their capacity to incorporate nonlinear interactions and behavioral heterogeneity makes them particularly effective for modeling housing market dynamics. In contrast to dynamic stochastic general equilibrium (DSGE) models, which typically rely on assumptions of rational behavior and agent homogeneity, agent-based models can capture emergent phenomena such as price fluctuations, speculative bubbles, and other complex dynamics that characterize housing markets.
Accordingly, the current study employed an agent-based modeling to examine the effects of different behaviors concerning the housing demand—including consumption-driven and investment-driven demand—as well as the impact of economic policies on housing prices. Specifically, the model simulated three main types of agents in the housing market: (1) owner–occupier households, who purchase housing primarily to satisfy their housing needs; (2) investors, who seek to generate returns through housing price appreciation; and (3) developers or builders, who operate in the construction sector with the objective of maximizing profits. Each type of agent makes decisions independently based on an optimization framework, with behavior guided by its own objectives.
The agent-based modeling helped examine how the heterogeneous behaviors of market agents influence housing prices and market volatility. Furthermore, the flexibility of the model enabled the simulation of a wide range of policy scenarios, allowing for a comprehensive assessment of their effects on housing market dynamics.
Materials and Methods
The present study used an agent-based modeling to examine the Iranian housing market. The data, obtained from the Central Bank of Iran and the Statistical Center of Iran, consisted of quarterly time-series observations spanning the period from 1998 to 2022. Python 3 and its associated libraries were used to estimate and analyze an economic model of the housing market. These tools supported both data analysis and model simulation. Moreover, initial model parameters (i.e., construction costs, financing costs, interest rates, and investors’ risk aversion) were derived from empirical data and evidence from related studies.
Model estimation and calibration were conducted within a heterogeneous agent modeling (HAM) framework grounded in agent-based models, which are well suited to capturing the complex, nonlinear, and asymmetric dynamics of the housing market. Unlike representative-agent models that assume homogeneous and fully rational behavior, the proposed model views agents as operating under bounded rationality and imperfect information.
Moreover, a genetic algorithm was employed to identify the optimal set of model parameters. By comparing the simulated outputs with the observed data using the standard Euclidean distance metric, the algorithm searches for the parameter combination that best reproduces the dynamics of housing prices in Iran. This computational framework provides a robust tool for analyzing market behavior and facilitates the simulation and evaluation of the effects of alternative policy interventions on housing price volatility.
Results and Discussion
In the analysis of systems of equations, particularly in the estimation of HAMs, model identification and convergence have consistently posed fundamental challenges because of the large number of parameters and the inherent complexity of HAM frameworks. To address these challenges, the current study simplified the system of equations by focusing on the key structural parameters, thereby enabling the estimation of a limited set of parameters.
Moreover, the key structural parameters of the housing price model were selected to minimize the distance between the empirically grounded parameters and those generated by the model. Given the dynamic relationship governing housing prices, the model incorporated four principal parameters: investors’ risk characteristics , the real demand factor , forecasting intensity , and construction costs . Once these four parameters are determined, the system of equations is fully specified. The estimated parameter values, together with the intensity of speculative expectations in investment behavior, are presented in the table.

Table 1. Estimated Values of Model Parameters





Estimated values


Model parameters




1/816059







0/431520







0/1947370







0/521691







0/2828







0/5655








Source: Research estimates
To evaluate the explanatory power of the model, a correlogram of the actual and fitted (ex post) values up to the tenth lag was employed. As shown in Figure 1, the model’s explanatory power declined as the lag length increased. Nevertheless, the correlation between the predicted and actual values remained above 78% through the tenth lag and reached 97% at the first lag.
Figure 1. Correlation Between Actual and Predicted Housing Prices at Different Lags

Source: Research estimates
Furthermore, based on the estimated values of , , and the model exhibited an unstable bubble regime, which is consistent with the observed conditions of the housing market over the period under study, ending in 2022. Moreover, the model provided a systemic framework for simulating and analyzing the effects of various policy interventions on housing price volatility. The current analysis examined several policy scenarios, including changes in construction costs and housing finance measures (e.g., reductions in down-payment requirements and the relaxation of mortgage repayment conditions).
Figure 2. Effects of Changes in the Loan-to-Value (LTV) Ratio Under Rising Housing Prices

Source: Research estimates
The analysis of housing loans and mortgage repayment costs indicated that the effectiveness of such policies would depend on their timing. Policy measures based on the ratio of affordable mortgage repayments have limited effectiveness in controlling the housing market, particularly in countries such as Iran, where housing loans account for only a small fraction of the total cost of housing during many periods, especially in recent years.
Unlike installment ratios, construction costs ( ) not only affected builders’ supply and the equilibrium price but also influenced housing prices through investors. The values and also depended on construction costs ( ). In the model, construction costs include land costs, material costs, and other related expenses, with land costs representing the largest component for builders.
Accordingly, when housing prices are rising, policymakers can effectively moderate the housing market by reducing builders’ costs—for example, by providing land at lower prices. Under this scenario, lower construction costs reduce the equilibrium housing price while increasing the share of investors who follow a mean-reversion strategy, thereby improving market rationality. At the same time, they reduce the sensitivity of speculative investors. As a result, housing prices can decline and converge to a new equilibrium level.
Similarly, in a declining market, reducing construction costs is a more effective policy measure because it leads to a stable new equilibrium price, thereby facilitating the market’s return to its fundamental level. This effect is illustrated in Figure 3.

Figure 3. Effects of Reduced Construction Costs on the Housing Market During Periods of Rising Prices

Source: Research estimates
Conversely, if the housing market is excessively overheated and experiencing a price bubble, increasing construction costs is not an appropriate policy for market stabilization. In some cases, this approach may even exacerbate market conditions. Higher construction costs raise the equilibrium price and increase the sensitivity of forward-looking investors—that is, they reduce —thereby making the housing market more unstable and increasing the likelihood of bubble formation or more severe market downturns.
Conclusion
The model presented in this study simulated speculative behavior among investors as a key driver of housing price fluctuations, demonstrating that such behavior could significantly influence housing price dynamics. The results indicated that, in the absence of speculative investors, the equilibrium housing price is determined solely by real demand and construction costs, representing the fundamental equilibrium price. However, in the presence of speculative behavior, housing prices may persistently deviate from this benchmark and may even exhibit explosive growth. The model further showed that greater forecasting intensity among investors could lead to larger deviations from the fundamental price and increased volatility in the housing market.
From a policy perspective, the model indicated that measures such as loan-to-value ratios and mortgage repayment requirements are not particularly effective tools for regulating the housing market during periods of boom or recession. Consistent with observed market data, changes in mortgage repayment ratios appear to have only limited short- and long-term effects on the housing market. In contrast, reducing construction costs can significantly decrease the sensitivity of speculative investors and enhance market rationality. Such measures help stabilize the fundamental housing price and support the recovery of the housing market.

Keywords

Housing
Agent-based Modeling
Speculative Behavior
Genetic Algorithm
Subjects
Amilon, H. (2005). Estimation of an adaptive stock market model with heterogeneous agents. Sveriges Riksbank Working Paper Series, No. 177, Sveriges Riksbank, Stockholm.
https://hdl.handle.net/10419/82413
Bahraminia, E., Abolhasani, A. & Ebrahimi, I. (2019). A new Keynesian DSGE model with a focus on the housing sector for Iran’s economy. The Journal Economic Policy, 10(20), 71-102. [In Persian].
Bolt, W., Demertzis, M., Diks, C., Hommes, C. & van der Leij, M. (2016). Identifying booms and busts in house prices under heterogeneous expectations. Journal of Economic Dynamics and Control. 103, 234-259. doi:10.1016/j.jedc.2019.04.003
Brock, W.A. & Hommes, C.H. (1997). A rational route to randomness. Econometrica, 65(5), 1059–1095. https://doi.org/10.2307/2171879
Brock, W.A. & Hommes, C.H. (1998). Heterogeneous beliefs and routes to chaos in a simple asset pricing model. Journal of Economic Dynamics and Control, 22(8–9), 1235-1274. doi: 10.1016/S0165-1889(98)00011-6.
Chang, X. & Lio, L. (2017). Characterizing rural household differentiation from the perspective of farmland transfer in eastern china using an agent based model. Springer Nature, 46(6),  875–886. DOI:10.1007/s10745-018-0035-6
Chen, S., Chang, Ch. & Du, Y. (2009). Agent-based economic models and econometrics. The Knowledge Engineering Review, 27(2), 187–219. doi: 10.1017/S0269888912000136
Chen, M.C. & Patel, K. (1998). House price dynamics and granger causality: An analysis of taipei new dwelling market. International Real Estate Review, Global Social Science Institute, 1(1), 101-126.
Chiarella, C., Iori, G. & Perelló, J. (2009). The impact of heterogeneous trading rules on the limit order book and order flows. Journal of Economic Dynamics and Control, 33(3), 525-537.
doi:10.1016/j.jedc.2008.08.001
Cho, M. (1998). House price dynamics: a survey of theoretical and empirical issues. mics: A Survey of Theoretical and Empirical Issues. Journal of Housing Research, Special Issue: House Price Indices: Policy, Business, and Research Applications, 7(2), 145-172.
https://www.jstor.org/stable/24832857
Cutler, D., Poterba, J. & Summers, L. (1990). Speculative dynamics and the role of feedback traders. NBER Working Paper.
Chiarella, C., He, X. &  Zwinkels, R.C.J. (2014). Heterogeneous expectations in asset pricing: Empirical evidence from the S&P500. Journal of Economic Behavior & Organization, 105,  1-16.
https://doi.org/10.1016/j.jebo.2014.03.003
Day, R. & Huang, W. (1990). Bulls, bears and market sheep. Journal of Economic Behavior & Organization, 14(3), 299–329.
https://doi.org/10.1016/0167-2681(90)90061-H
Dieci, R. & Westerhoff, F. (2009). A simple model of a specul (Lux, Thomas, 1998) ative housing market. BERG Working Paper Series on Government and Growth, 62. https://hdl.handle.net/10419/38760
DiPasquale, D. & Wheaton, W. (1992). The markets for real estate assets and space: A conceptual framework. Real Estate Economics, American Real Estate and Urban Economics Association, 20(2), 181-198. DOI: 10.1111/1540-6229.00579
Föllmer, H., Horst, U. & Kirman, A., (2004). Equilibria in financial markets with heterogeneous agents: a probabilistic perspective. Journal of Mathematical Economics, 41(1-2), 123–155.
doi:10.1016/j.jmateco.2004.08.001
Gamal, Y., Elsenbroich, C., Gilbert, N., Heppenstall, A. & Zia, K. (2024). A behavioural agent-based model for housing markets: impact of financial shocks in the UK. Journal of Artificial Societies and Social Simulation, 27(4), 1-5.
Gholizade, A.A. & Noroozonejad, M. (2019). Dynamics of housing prices and economic fluctuations in iran with the approach of dynamic stochastic general equilibrium (DSGE). The Journal of Economic Modeling Research, 10(36), 37-74. 10.29252/jemr.9.36.37. [In Persian].
Gholizadeh, A.A., Manochehri, S. & Jafariseresht, D. (2023). Time-varying effects of factors influencing speculation in Iran's housing market: state-space models. Econometric Modeling, 7(4), 119-142.
10.22075/jem.2023.29886.1810. [In Persian].
Gholizadeh, A.A. & Samadipour, S. (2023). The effect of heterogeneous behavior of investors of housing sector on inflation from the housing price channel. The Journal of Econometric Modeling, 8(3), 163-188. 10.22075/jem.2023.31008.1853. [In Persian].
Gibbs, C.G., Hambur, J. & Nodari, G. (2018). DSGE Reno: Adding a housing block to a small open economy model. UNSW Sydney Economic Research Department, Reserve Bank of Australia,April.
He, X. & Li, Y. (2015). Testing of a market fraction model and power-law behaviour in the DAX 30. Journal of Empirical Finance, 31, 1–17.doi: 10.1016/j.jempfin.2015.01.001
He, Y. & Xia, F. (2020). Heterogeneous traders, house prices and healthy urban housing market: A DSGE model based on behavioral economics. Habitat International, 96. 10.1016/j.habitatint.2019.102085.
He, X.Z., Li, K., Wei, J. & Zheng, M. (2009). Market stability switches in a continuous-time financial market with heterogeneous beliefs. Economic Modelling, 26(6), 1432–1442.
https://doi.org/10.1016/j.econmod.2009.07.016
Hojjat, S., Mehrara, M. & Taiebnia, A. (2021). The impact of monetary and fiscal policies on bubbles in the real estate sector of the iranian economy: an agent-based approach. Journal of Economic Research, 56(2), 257-289. 10.22059/jte.2021.330940.1008547 [In Persian].
Hong, Y. & Li, Y. (2019). Housing prices and investor sentiment dynamics: Evidence from China using a wavelet approach. Finance Research Letters, 35. https://doi.org/10.1016/j.frl.2019.09.015.
 Kabundi, A., Chaling, E. & Some, M. (2015). Monetary policy and heterogeneous inflation expectations in South Africa. Economic Modelling, 45,  109-117.
https://doi.org/10.1016/j.econmod.2014.11.002
Keyfarokhi, I., Akbari, N., Farahmand, S. & Asgary, A. (2020). Simulation of the housing rental market using agent-based modeling case study: district 6 of Isfahan city. Journal of Economic Research, 55(1), 135-165. 10.22059/jte.2020.282224.1008175. [In Persian].
Li, Y., Donkers, B. & Melenberg, B. (2010). Econometric analysis of microscopic simulation models. Quantitative Finance, 10(10), 1187–1201. https://doi.org/10.1080/14697680903460176
Lux, T. (1998). The socio-economic dynamics of speculative markets: interacting agents, chaos, and the fat tails of return distributions. Journal of Economic Behavior & Organization, 33(2), 143-165. https://doi.org/10.1016/S0167-2681(97)00088-7
Mahdavipoor, M., Shirmard Ahmad Abad, H. & Mortezania, H. (2024). Interpretive-structural modeling of behavioral biases of housing sector investors. Strategic Research on Budget and Public Finance, 5(1), 79-101. [In Persian].
Malek, H., Delangizan, S. & Almasi, M. (2022). Investigating the role of housing finance in iranian business cycles, DSGE approach. The Quarterly Journal of Quantitative Economics, 19(3), 63-92.
doi: 10.22055/jqe.2021.32274.2201. [In Persian].
Malpezzi, S. & Wachter, S.M. (2005). The role of speculation in real estate cycles. Journal of Real Estate Literature, 13(2), 143-164. https://www.jstor.org/stable/44103516.
Manoochehri, S., Gholizade, A. & Jafariseresht, D. (2024). the dynamics of speculation in Iran's housing market. Journal of Urban Economics and Management, 12(46), 25-47. URL: http://iueam.ir/article-1-2102-fa.html. [In Persian].
Mérő, B., Borsos, A., Hosszú, Z., Oláh, Z. & Vágó, N. (2023). A high-resolution, data-driven agent-based model of the housing market. Journal of Economic Dynamics and Control, 155.
doi: 10.1016/j.jedc.2023.104738.
Mousavi, M., Khezri, A., Raghfar, H. & Sangari Mohazab, K. (2023). The simulation of housing price in Tehran: an spatial agent based approach. Economic Research, 58(1), 151-183. 10.22059/jte.2023.93460. [In Persian].
Napoletano, M., Gaffard, J. & Babutsidze, Z. (2014). Agent based models a new tool for economic and policy analysis. Hal Open Science, 3, 1-15. https://sciencespo.hal.science/hal-01070338v1.
Rauf, M.A. & Weber, O. (2022). Housing sustainability: the effects of speculation and property taxes on house prices within and beyond the jurisdiction. Sustainability, 14(12).
https://doi.org/10.3390/su14127496
Samadipour, S., Gholizadeh, A.A. & Sepehrdoust, H. (2023). Investigating the impact of behavioral factors on the iranian housing price. The Quarterly Journal of Applied Economic Studies in Iran, 12(46), 241-274. https://dx.doi.org/10.22084/AES.2023.27055.3531. [In Persian].
Shi, S., Rahman, A. & Wang, B.Z. (2020). Australian housing market booms: fundamentals or speculation? Economic Record, 96(315), 381-401. https://doi.org/10.1111/1475-4932.12553
Smith, S.J. (2011). Home price dynamics: a behavioural economy? Housing, Theory and Society, 28(3), 236-261.
DOI:10.1080/14036096.2011.599179.
Tarne, R. & Bezemer, D. (2025). Roof or real estate? An agent-based model of housing affordability in the Netherlands. Structural Change and Economic Dynamics, 72, 163-178.
https://doi.org/10.1016/j.strueco.2024.11.014
Wang, Y., Xu, F. & Hu, A. (2013). Impact of heterogeneous beliefs and short sale constraints on security issuance decisions. Economic Modelling, 30, 539–545. https://doi.org/10.1016/j.econmod.2012.09.051
Zheng, M., Wang, H., Wang, C. & Wang, S. (2017). Speculative behavior in a housing market: Boom and bust. Economic Modelling, 61, 50-64. https://doi.org/10.1016/j.econmod.2016.11.021

  • Receive Date 17 July 2025
  • Revise Date 06 February 2026
  • Accept Date 17 December 2025
  • First Publish Date 22 December 2025

Home

editorial board

Guide for Authors

Submit Manuscript

Contact Us