Working Papers

Estimating Nonlinear Heterogeneous Agent Models with Neural Networks

R&R Econometrica (2nd round)
Working Papers
We leverage recent advancements in machine learning to develop an integrated method to solve globally and estimate models featuring agent heterogeneity, nonlinear constraints, and aggregate uncertainty. Using simulated data, we show that the proposed method accurately estimates the parameters of a nonlinear Heterogeneous Agent New Keynesian (HANK) model with a zero lower bound (ZLB) constraint. We further apply our method to estimate this HANK model using U.S. data. In the estimated model, the interaction between the ZLB constraint and idiosyncratic income risks emerges as a key source of aggregate output volatility.
PDF Code
    Presented at
  • The Society for Nonlinear Dynamics and Econometrics (SNDE) Symposium (March 2022)
  • The Society for Economic Measurement (SEM) Annual Conference (August 2022)
  • The European Economic Association (EEA) and European meeting of the Econometric Society (ESEM) Conference (August 2022)
  • Conference on Non-traditional Data, Machine Learning and Natural Language Processing in Macroeconomics at Sveriges Riksbank (October 2022)
  • Midwest Macro Meeting (November 2022)
  • ASSA (January 2023)
  • CEF (July 2023)
  • NBER Summer Institute (July 2023)
  • DSE (August 2023)
  • SED (June 2024)
  • ECB (October 2024)
  • Goethe University Frankfurt, Numerical Methods in Macroeconomic (October 2024)

Beyond Averages: Heterogeneous Effects of Monetary Policy in a HANK Model for the Euro Area

Working Papers
We introduce an estimated medium scale Heterogeneous-Agent New Keynesian model for forecasting and policy analysis in the Euro Area and discuss the applications of this type of models in central banks, focusing on two main exercises. First, we examine an alternative scenario for monetary policy during the early 2020s inflationary episode, showing that earlier hikes in interest rates would have affected more strongly households at the lower end of the wealth distribution, whose consumption our model suggests was already depressed relative to the rest of the population. To provide intuition for this result, we introduce a new decomposition of the effects of monetary policy on consumption across the wealth distribution. Second, we show that introducing heterogeneous households does not come at the cost of forecasting accuracy by comparing the performance of our model to its exact representative-agent counterpart and demonstrating nearly identical results in predicting key aggregate variables.
ECB WP
    Presented at
  • ECB (June 2025)

Generative Economic Modeling

Working Papers
We introduce a novel approach for solving quantitative economic models: generative economic modeling. Our method combines neural networks with conventional solution techniques. Specifically, we train neural networks on simplified versions of the economic model to approximate the complete model’s dynamic behavior. Relying on these less complex sub-models circumvents the curse of dimensionality, allowing the use of well-established numerical methods. We demonstrate our approach across settings with analytical characterizations, nonlinear dynamics, and heterogeneous agents, employing asset pricing and business cycle models. Finally, we solve a high-dimensional HANK model with financial frictions to highlight how aggregate risk amplifies the precautionary motive.
PDF Code
    Presented at
  • Dynare Conference (June 2025)

The Perils of Narrowing Fiscal Spaces

Working Papers
When public debt is elevated, the fiscal cost of fighting inflation rises sharply, as interest rate hikes increase government interest expenditures. We formalize this mechanism in a nonlinear New Keynesian model with a state-dependent fiscal constraint on monetary policy. High debt may dampen the monetary response to inflation, generating an inflationary bias even though government debt remains fully fiscally backed. The interaction between high debt and inflationary cost-push shocks makes the fiscal limit more likely to bind, amplifying inflation. In demand-driven downturns, the fiscal constraint may become more restrictive than the zero lower bound, forcing the central bank to either print money to purchase excess debt or accept fiscal dominance.
PDF

Work in Progress

Sequence-Space Jacobian meets Deep Learning: Exploiting the Random Walk for HANK

Work in Progress
This paper introduces an innovative approach integrating deep learning techniques with sequence-space Jacobian methods to enhance Bayesian estimation in heterogeneous agent New-Keynesian (HANK) models. By employing a deep neural network as a surrogate for the posterior, we aim to accelerate the Bayesian estimation process significantly. This network is trained on a dataset comprising true model likelihoods generated through a parallel Metropolis-Hastings algorithm. Our method uniquely leverages all generated draws, including both accepted and rejected ones, thereby ensuring a thorough exploration of the parameter space. This strategy not only alleviates the computational burden traditionally associated with Bayesian estimation but also demonstrates remarkable efficiency in estimating parameters that necessitate the resolution of the model's steady state and the recalculation of Jacobians. Our work stands at the frontier of integrating advanced computational techniques with economic modeling, promising substantial advancements in estimating and understanding complex heterogeneous agent models.
Slides
    Presented at
  • CEF (June 2024)
  • SEM (August 2024)

Resting

Backpropagating Through Heterogeneous Agent Models

Resting
This paper explores applications of the backpropagation algorithm on heterogeneous agent models. In addition, I clarify the connection between deep learning and dynamic structural models by showing how a standard value function iteration algorithm can be viewed as a recurrent convolutional neural network. As a result, many advances in the field of machine learning can carry over to economics. This in turn makes the solution and estimation of more complex models feasible.

Limits on Mortgage Lending

Resting
This paper aims to study the impact of macroprudential limits on mortgage lending in a heterogeneous agent life-cycle model with incomplete markets, long-term mortgages, and defaults. Using data from the Household Finance and Consumption Survey, the model is calibrated for the German economy. I consider the effects of four policy instruments: loan-to-value limit, debt-to-income limit, payment-to-income limit, and maximum maturity. I find that their effect on the homeownership rate is fairly modest. Only the loan-to-value limit significantly reduces the homeownership rate among young households. At the same time, it has the most significant positive welfare effect
No matching items