Hazhir Rahmandad
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Hazhir Rahmandad has a total of 88 co-authors. Here are the top co-authors by number of shared publications:
- Mohammad S. Jalali
Affiliation MGH Institute for Technology Assessment, Harvard Medical School, Boston, Massachusetts, United States of America
8 shared publications - John Sterman
Massachusetts Institute of Technology (MIT) - Sloan School of Management
6 shared publications - Navid Ghaffarzadegan
Department of Industrial and Systems Engineering Virginia Tech Blacksburg Virginia USA
5 shared publications - Alice S. Ammerman
University of North Carolina at Chapel Hill
3 shared publications - Sally Lawrence Bullock
Department of Nutrition, Gillings School of Global Public Health, UNC Center for Health Promotion and Disease Prevention (a CDC Prevention Research Center), University of North Carolina at Chapel Hill, NC, USA
3 shared publications Tse Yang Lim
Massachusetts Institute of Technology Cambridge MA USA
3 shared publications
Latest Publications
- Open Access
- 30 March 2026
- Published by Wiley in Risk Analysis
- Vol. 46 (4) , e70228
Faster Uptake, Slower Let‐Down: Asymmetric Community Responses to Changing Risks During a Pandemic
From how public opinion responds to economic outcomes to how risk perception shapes social interactions during a pandemic, many important social processes involve the assimilation of information to form opinions and perceptions, which in turn guide individual and societal actions. While such perception delays are well recognized, empirically identifying them is nontrivial. We study this problem in the context of human responses to disease dynamics, mediated by public risk perception. Public risk perception changes through an information diffusion process with significant delays, where perception adjustment delay may differ depending on whether the risk is increasing or decreasing. Despite these complexities, most models assume either fixed‐delay structures or exponential structures with symmetric delay periods. First, using synthetic data (where ground truth is known), we show that incorrect delay structures and the assumption of symmetric delay periods can lead to biased and misleading estimates. We then explore alternative approaches to identify more appropriate delay structures that can overcome these challenges. Second, we apply these asymmetric delay structures to state‐level US COVID‐19 disease and mobility data to demonstrate how estimates of public sensitivity to mortality depend on the assumed delay structure. Our results provide evidence that during the pandemic, human risk perception adjusted with asymmetric delays to changing risk: people perceived rising risks more quickly than they perceived declining risks.
- 30 March 2026
- Published by Wiley in Risk Analysis
- Vol. 46 (4) , e70228
Open AccessFaster Uptake, Slower Let‐Down: Asymmetric Community Responses to Changing Risks During a Pandemic
From how public opinion responds to economic outcomes to how risk perception shapes social interactions during a pandemic, many important social processes involve the assimilation of information to form opinions and perceptions, which in turn guide individual and societal actions. While such perception delays are well recognized, empirically identifying them is nontrivial. We study this problem in the context of human responses to disease dynamics, mediated by public risk perception. Public risk perception changes through an information diffusion process with significant delays, where perception adjustment delay may differ depending on whether the risk is increasing or decreasing. Despite these complexities, most models assume either fixed‐delay structures or exponential structures with symmetric delay periods. First, using synthetic data (where ground truth is known), we show that incorrect delay structures and the assumption of symmetric delay periods can lead to biased and misleading estimates. We then explore alternative approaches to identify more appropriate delay structures that can overcome these challenges. Second, we apply these asymmetric delay structures to state‐level US COVID‐19 disease and mobility data to demonstrate how estimates of public sensitivity to mortality depend on the assumed delay structure. Our results provide evidence that during the pandemic, human risk perception adjusted with asymmetric delays to changing risk: people perceived rising risks more quickly than they perceived declining risks.
- Open Access
- 1 April 2025
- Published by Wiley in System Dynamics Review
- Vol. 41 (2)
Beyond Least Squares: Estimation of Dynamic Models With Alternative Likelihoods and Kalman Filtering
From business to healthcare and operations to strategy, grounding system dynamics models in data is indispensable for theory and practice. However, formal estimation is difficult due to incomplete data, model mis‐specification, process noise, and measurement error. This complexity has limited the quantity and quality of formal estimation. We argue that comparing generic and easy‐to‐apply estimation methods for common models is fruitful for identifying methods that work well for SD practitioners. Using the classical SEIR model, we compare standard least squares against maximum likelihood estimators including variance‐scaled Gaussian, log Gaussian, Poisson, and negative binomial estimators, and assess the value of (extended) Kalman filtering. Under different assumptions about data availability and noise, we find that least squares, log Gaussian, and scaled Gaussian likelihoods perform poorly in estimating confidence intervals. The negative binomial and Kalman filtering with variance scaling and auto‐correlated process noise are promising across different setups. Implications for modelers are discussed.
- 1 April 2025
- Published by Wiley in System Dynamics Review
- Vol. 41 (2)
Open AccessBeyond Least Squares: Estimation of Dynamic Models With Alternative Likelihoods and Kalman Filtering
From business to healthcare and operations to strategy, grounding system dynamics models in data is indispensable for theory and practice. However, formal estimation is difficult due to incomplete data, model mis‐specification, process noise, and measurement error. This complexity has limited the quantity and quality of formal estimation. We argue that comparing generic and easy‐to‐apply estimation methods for common models is fruitful for identifying methods that work well for SD practitioners. Using the classical SEIR model, we compare standard least squares against maximum likelihood estimators including variance‐scaled Gaussian, log Gaussian, Poisson, and negative binomial estimators, and assess the value of (extended) Kalman filtering. Under different assumptions about data availability and noise, we find that least squares, log Gaussian, and scaled Gaussian likelihoods perform poorly in estimating confidence intervals. The negative binomial and Kalman filtering with variance scaling and auto‐correlated process noise are promising across different setups. Implications for modelers are discussed.
- Open Access
- 21 January 2025
- Published by Wiley in System Dynamics Review
- Vol. 41 (1)
Incorporating Deep Learning Into System Dynamics: Amortized Bayesian Inference for Scalable Likelihood‐Free Parameter Estimation
Estimating parameters and their credible intervals for complex system dynamics models is challenging but critical to continuous model improvement and reliable communication with an increasing fraction of audiences. The purpose of this study is to integrate Amortized Bayesian Inference (ABI) methods with system dynamics. Utilizing Neural Posterior Estimation (NPE), we train neural networks using synthetic data (pairs of ground truth parameters and outcome time series) to estimate parameters of system dynamics models. We apply this method to two example models: a simple Random Walk model and a moderately complex SEIRb model. We show that the trained neural networks can output the posterior for parameters instantly given new unseen time series data. Our analysis highlights the potential of ABI to facilitate a principled, scalable, and likelihood‐free inference workflow that enhance the integration of models of complex systems with data. Accompanying code streamlines application to diverse system dynamics models.
- 21 January 2025
- Published by Wiley in System Dynamics Review
- Vol. 41 (1)
Open AccessIncorporating Deep Learning Into System Dynamics: Amortized Bayesian Inference for Scalable Likelihood‐Free Parameter Estimation
Estimating parameters and their credible intervals for complex system dynamics models is challenging but critical to continuous model improvement and reliable communication with an increasing fraction of audiences. The purpose of this study is to integrate Amortized Bayesian Inference (ABI) methods with system dynamics. Utilizing Neural Posterior Estimation (NPE), we train neural networks using synthetic data (pairs of ground truth parameters and outcome time series) to estimate parameters of system dynamics models. We apply this method to two example models: a simple Random Walk model and a moderately complex SEIRb model. We show that the trained neural networks can output the posterior for parameters instantly given new unseen time series data. Our analysis highlights the potential of ABI to facilitate a principled, scalable, and likelihood‐free inference workflow that enhance the integration of models of complex systems with data. Accompanying code streamlines application to diverse system dynamics models.
- 1 January 2024
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Organization Science
- Vol. 35 (1) , 52-70
How Does Network Structure Impact Socially Reinforced Diffusion?
How does network structure impact the speed and reach of social contagions? The current view holds that random links facilitate “simple” contagion, but when agents require multiple reinforcements for “complex” adoption, clustered networks are better conduits of social influence. We show that in complex contagion, even low probabilities of adoption upon a single contact would activate an exponential contagion process that tilts the balance in favor of random networks. On the other hand, underappreciated but critical to the race between random and clustered networks is how long agents engage with contagion. Switching back to prior practice and the inactivation of senders and especially receivers shorten the window of engagement for convincing distant contacts and weaken the reach of diffusion on random networks. We propose a simplified framework where clustering primarily enables contagion when repetition matters and receivers lose interest quickly; otherwise, diffusion, simple or complex, is faster on random networks than clustered ones. These mechanisms can inform designing social networks, structuring groups, and seeding of ideas and innovations at a time when the increasing inflow of content from various media limits actors’ engagement with each item, whereas expanding network size and connections speeds up diffusion through distant contacts. Supplemental Material: The e-companion is available at https://doi.org/10.1287/orsc.2023.1658 .
- 1 January 2024
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Organization Science
- Vol. 35 (1) , 52-70
How Does Network Structure Impact Socially Reinforced Diffusion?
How does network structure impact the speed and reach of social contagions? The current view holds that random links facilitate “simple” contagion, but when agents require multiple reinforcements for “complex” adoption, clustered networks are better conduits of social influence. We show that in complex contagion, even low probabilities of adoption upon a single contact would activate an exponential contagion process that tilts the balance in favor of random networks. On the other hand, underappreciated but critical to the race between random and clustered networks is how long agents engage with contagion. Switching back to prior practice and the inactivation of senders and especially receivers shorten the window of engagement for convincing distant contacts and weaken the reach of diffusion on random networks. We propose a simplified framework where clustering primarily enables contagion when repetition matters and receivers lose interest quickly; otherwise, diffusion, simple or complex, is faster on random networks than clustered ones. These mechanisms can inform designing social networks, structuring groups, and seeding of ideas and innovations at a time when the increasing inflow of content from various media limits actors’ engagement with each item, whereas expanding network size and connections speeds up diffusion through distant contacts. Supplemental Material: The e-companion is available at https://doi.org/10.1287/orsc.2023.1658 .
- 1 November 2023
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Organization Science
- Vol. 34 (6) , 2392-2414
Delays Impair Learning and Can Drive Convergence to Inefficient Strategies
With so many possible choices, why do managers adopt the strategies they do? We identify delays between adopting a strategy and observing the full implications of that choice as a critical factor influencing strategic choices. Using a simulation of a service firm, we conduct two behavioral experiments to investigate how delays interact with outcome uncertainty to shape learning, strategy adaptation, and performance outcomes. Two mechanisms emerge from how different subject groups perceive, react to, and learn in the presence of delayed feedback and uncertainty. First, when multiple viable strategies exist, longer delays lead both general participants and experienced managers toward alternatives that have rapid returns. When those alternatives are suboptimal, delays may strengthen convergence to inefficient strategies. Second, delays and uncertainty may also induce learners to persist with their a priori strategies. Managers show larger confidence in their priors and thus underperform general participants when the underlying task structure diverges from those priors. Both mechanisms can undermine performance. Moreover, delays and uncertainty may reduce heterogeneity in strategies and performance in more dynamic, uncertain environments, leading to convergence as tasks grow more complex and where decision makers possess similar priors. History: This paper has been accepted for the Organization Science Special Issue on Experiments in Organizational Theory. Funding: Funding from the Massachusetts Institute of Technology and UNSW Business School supported this research. Supplemental Material: The online appendices are available at https://doi.org/10.1287/orsc.2020.1405 .
- 1 November 2023
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Organization Science
- Vol. 34 (6) , 2392-2414
Delays Impair Learning and Can Drive Convergence to Inefficient Strategies
With so many possible choices, why do managers adopt the strategies they do? We identify delays between adopting a strategy and observing the full implications of that choice as a critical factor influencing strategic choices. Using a simulation of a service firm, we conduct two behavioral experiments to investigate how delays interact with outcome uncertainty to shape learning, strategy adaptation, and performance outcomes. Two mechanisms emerge from how different subject groups perceive, react to, and learn in the presence of delayed feedback and uncertainty. First, when multiple viable strategies exist, longer delays lead both general participants and experienced managers toward alternatives that have rapid returns. When those alternatives are suboptimal, delays may strengthen convergence to inefficient strategies. Second, delays and uncertainty may also induce learners to persist with their a priori strategies. Managers show larger confidence in their priors and thus underperform general participants when the underlying task structure diverges from those priors. Both mechanisms can undermine performance. Moreover, delays and uncertainty may reduce heterogeneity in strategies and performance in more dynamic, uncertain environments, leading to convergence as tasks grow more complex and where decision makers possess similar priors. History: This paper has been accepted for the Organization Science Special Issue on Experiments in Organizational Theory. Funding: Funding from the Massachusetts Institute of Technology and UNSW Business School supported this research. Supplemental Material: The online appendices are available at https://doi.org/10.1287/orsc.2020.1405 .
- 8 October 2023
- Published by Wiley in System Dynamics Review
- Vol. 39 (4) , 453-455
Call for submissions to the 2024 ISDC
- 8 October 2023
- Published by Wiley in System Dynamics Review
- Vol. 39 (4) , 453-455
Call for submissions to the 2024 ISDC
- Open Access
- 24 August 2022
- Published by Wiley in System Dynamics Review
- Vol. 38 (4) , 329-353
Quantifying the COVID‐19 endgame: Is a new normal within reach?
Eradication of COVID‐19 is out of reach. Are we close to a “new normal” in which people can leave behind restrictive non‐pharmaceutical interventions (NPIs) yet face a tolerable burden of disease? The answer depends on the ongoing risks versus communities' tolerance for those risks. Using a detailed model of the COVID‐19 pandemic spanning 93 countries, we estimate the biological and behavioral factors determining the risks and responses, and project the likely course of COVID‐19. Infection fatality rates have fallen significantly due to vaccination, prior infections, better treatments, and the less severe Omicron variant. Yet based on their estimated tolerance for deaths, most nations are not ready to live with COVID‐19 without any NPIs. Across the world the increased transmissibility of Omicron, combined with the decay of immunity, leads to repeated episodes of reinfections, hospitalizations, and deaths, complicating the emergence of a new normal in many nations. © 2022 The Authors. System Dynamics Review published by John Wiley & Sons Ltd on behalf of System Dynamics Society.
- 24 August 2022
- Published by Wiley in System Dynamics Review
- Vol. 38 (4) , 329-353
Open AccessQuantifying the COVID‐19 endgame: Is a new normal within reach?
Eradication of COVID‐19 is out of reach. Are we close to a “new normal” in which people can leave behind restrictive non‐pharmaceutical interventions (NPIs) yet face a tolerable burden of disease? The answer depends on the ongoing risks versus communities' tolerance for those risks. Using a detailed model of the COVID‐19 pandemic spanning 93 countries, we estimate the biological and behavioral factors determining the risks and responses, and project the likely course of COVID‐19. Infection fatality rates have fallen significantly due to vaccination, prior infections, better treatments, and the less severe Omicron variant. Yet based on their estimated tolerance for deaths, most nations are not ready to live with COVID‐19 without any NPIs. Across the world the increased transmissibility of Omicron, combined with the decay of immunity, leads to repeated episodes of reinfections, hospitalizations, and deaths, complicating the emergence of a new normal in many nations. © 2022 The Authors. System Dynamics Review published by John Wiley & Sons Ltd on behalf of System Dynamics Society.
- Open Access
- 23 August 2022
- Published by Wiley in System Dynamics Review
- Vol. 38 (3) , 246-263
Behavioral responses to risk promote vaccinating high‐contact individuals first
How should communities prioritize COVID‐19 vaccinations? Prior studies found that prioritizing the elderly and most vulnerable minimizes deaths. However, prior research has ignored how behavioral responses to risk of disease endogenously change transmission rates. We show that incorporating risk‐driven behavioral responses enhances fit to data and may change prioritization to vaccinating high‐contact individuals. Behavioral responses matter because deaths grow exponentially until communities are compelled to reduce contacts, with deaths stabilizing at levels that oblige higher‐contact groups to sufficiently cut their interactions and slow transmissions. More lives may be saved by vaccinating and taking those high‐contact groups out of transmission chains earlier because the remaining groups will take more precautions while waiting for their turn for vaccination. These findings are especially important considering the need for further vaccination in many countries, the emergence of new variants, and the expected challenge of distributing new vaccines in the coming months and years. © 2022 The Author. System Dynamics Review published by John Wiley & Sons Ltd on behalf of System Dynamics Society.
- 23 August 2022
- Published by Wiley in System Dynamics Review
- Vol. 38 (3) , 246-263
Open AccessBehavioral responses to risk promote vaccinating high‐contact individuals first
How should communities prioritize COVID‐19 vaccinations? Prior studies found that prioritizing the elderly and most vulnerable minimizes deaths. However, prior research has ignored how behavioral responses to risk of disease endogenously change transmission rates. We show that incorporating risk‐driven behavioral responses enhances fit to data and may change prioritization to vaccinating high‐contact individuals. Behavioral responses matter because deaths grow exponentially until communities are compelled to reduce contacts, with deaths stabilizing at levels that oblige higher‐contact groups to sufficiently cut their interactions and slow transmissions. More lives may be saved by vaccinating and taking those high‐contact groups out of transmission chains earlier because the remaining groups will take more precautions while waiting for their turn for vaccination. These findings are especially important considering the need for further vaccination in many countries, the emergence of new variants, and the expected challenge of distributing new vaccines in the coming months and years. © 2022 The Author. System Dynamics Review published by John Wiley & Sons Ltd on behalf of System Dynamics Society.
- Open Access
- 19 May 2022
- Published by Public Library of Science (PLoS) in PLOS Computational Biology
- Vol. 18 (5) , e1010100
Enhancing long-term forecasting: Learning from COVID-19 models
While much effort has gone into building predictive models of the COVID-19 pandemic, some have argued that early exponential growth combined with the stochastic nature of epidemics make the long-term prediction of contagion trajectories impossible. We conduct two complementary studies to assess model features supporting better long-term predictions. First, we leverage the diverse models contributing to the CDC repository of COVID-19 USA death projections to identify factors associated with prediction accuracy across different projection horizons. We find that better long-term predictions correlate with: (1) capturing the physics of transmission (instead of using black-box models); (2) projecting human behavioral reactions to an evolving pandemic; and (3) resetting state variables to account for randomness not captured in the model before starting projection. Second, we introduce a very simple model, SEIRb, that incorporates these features, and few other nuances, offers informative predictions for as far as 20-weeks ahead, with accuracy comparable with the best models in the CDC set. Key to the long-term predictive power of multi-wave COVID-19 trajectories is capturing behavioral responses endogenously: balancing feedbacks where the perceived risk of death continuously changes transmission rates through the adoption and relaxation of various Non-Pharmaceutical Interventions (NPIs).
- 19 May 2022
- Published by Public Library of Science (PLoS) in PLOS Computational Biology
- Vol. 18 (5) , e1010100
Open AccessEnhancing long-term forecasting: Learning from COVID-19 models
While much effort has gone into building predictive models of the COVID-19 pandemic, some have argued that early exponential growth combined with the stochastic nature of epidemics make the long-term prediction of contagion trajectories impossible. We conduct two complementary studies to assess model features supporting better long-term predictions. First, we leverage the diverse models contributing to the CDC repository of COVID-19 USA death projections to identify factors associated with prediction accuracy across different projection horizons. We find that better long-term predictions correlate with: (1) capturing the physics of transmission (instead of using black-box models); (2) projecting human behavioral reactions to an evolving pandemic; and (3) resetting state variables to account for randomness not captured in the model before starting projection. Second, we introduce a very simple model, SEIRb, that incorporates these features, and few other nuances, offers informative predictions for as far as 20-weeks ahead, with accuracy comparable with the best models in the CDC set. Key to the long-term predictive power of multi-wave COVID-19 trajectories is capturing behavioral responses endogenously: balancing feedbacks where the perceived risk of death continuously changes transmission rates through the adoption and relaxation of various Non-Pharmaceutical Interventions (NPIs).
- Open Access
Behavioral Responses to Risk Promote Vaccinating High-contact Individuals First
If COVID-19’s reproduction number was constant, vaccinating elderly first minimized deaths. However, incorporating risk-driven behavior/policy changes enhances fit to data and prioritizes vaccinating high-contact individuals. Deaths grow exponentially until people are compelled to reduce contacts, stabilizing at levels obliging higher-contact groups to sufficiently cut interactions. Vaccinating those groups out of transmission saves lives and speeds everybody’s return to normal life.
Open AccessBehavioral Responses to Risk Promote Vaccinating High-contact Individuals First
If COVID-19’s reproduction number was constant, vaccinating elderly first minimized deaths. However, incorporating risk-driven behavior/policy changes enhances fit to data and prioritizes vaccinating high-contact individuals. Deaths grow exponentially until people are compelled to reduce contacts, stabilizing at levels obliging higher-contact groups to sufficiently cut interactions. Vaccinating those groups out of transmission saves lives and speeds everybody’s return to normal life.