Matthias Rottner

🚨How can we use machine learning techniques to solve and estimate HANK models?🚨 We updated our paper on estimating nonlinear HANK with neural networks. Now with a first set of example codes for our method! Check out github.com/tseep/estimati… @HannoKase and @LeonardoMelosi

Super interesting! "The perils of narrowing fiscal spaces" by Hanno Kase, Leonardo Melosi, Sebastian Rast, and Matthias Rottner. "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." bis.org/publ/work1328.…

New paper out 📉📄 Large public debt can seriously constrain central banks’ ability to fight inflation. Comments welcome! Coauthored with @HannoKase @MatthiasRottner @seb_rast.

Rising government debt makes rate hikes fiscally costly and may intensify pressure on central banks. A new model with an endogenous upper bound on the interest rate coming from debt service constraints shows how an easing bias can fuel persistent inflation bis.org/publ/work1328.…

Using generative economic modelling, a novel method to solve economic models, we examine whether aggregate risk amplifies economic dynamics in a macroeconomic model with heterogeneous agents and financial frictions. bit.ly/3M3hRXL

The BIS Multisector Model, a new macroeconomic model featuring a detailed production network and accompanied by a ready-to-use toolbox, analyses economic dynamics and monetary policy reactions in more than 80 economies. bit.ly/48f0UT8

How does the inflation environment shape the effects of monetary policy? High-frequency data show that, in a high-inflation regime, monetary policy shocks strongly amplify labour income inequality and aggregate consumption bis.org/publ/work1271.…

I have a new working paper on solving complex life-cycle models with deep learning. Paper: drive.google.com/file/d/1txCSkw… Python toolbox: github.com/NumEconCopenha…

Climate Minsky moments: Why they may matter less than expected Matthias Kaldorf @matthiaskaldorf @bundesbank, Matthias Rottner @MatthiasRottner @BIS_org ow.ly/2TEe50UTPYW

How does ambitious climate policy affect financial stability and how relevant are such "Climate Minsky Moments" for the macroeconomiy and welfare? @MatthiasRottner and I provide an answer through the lenses of a quantitative nonlinear DSGE model: cepr.org/voxeu/columns/…

Can artificial intelligence be used to solve and estimate HANK models? A neural-network-based approach to estimate a non-linear HANK model highlights the crucial interaction between aggregate and idiosyncratic risk bis.org/publ/work1241.…

Excited to see our paper now available as a BIS Working Paper as well!📄✨ If you are interested in our method, we have also released an initial set of example codes—check them out! github.com/tseep/estimati…

Growing geopolitical disruptions, climate change and a bumpy transition to green energy threaten to make commodity price shifts larger and more frequent going forward. Read more here bis.org/publ/bisbull96…

Qingyuan prepared his own modified version where he presents a simple and detailed walk through of the first part of their code, going step-by-step through some of the more complicated pieces. Check it out here! github.com/QingyuanFang/K…

Last year, the JHU economics ML reading group worked through "Estimating Nonlinear Heterogeneous Agent Models with Neural Networks" from @MatthiasRottner, @HannoKase, and @LeonardoMelosi. Today @Qingyuan_Fang walked us through the authors' code: github.com/tseep/estimati…

@pablo_guerron @YuchengYang1993 @uzh_df @SFI_CH @HannoKase @LeonardoMelosi Thanks a lot!

@YuchengYang1993 @MatthiasRottner @uzh_df @SFI_CH @HannoKase @LeonardoMelosi Great paper. Learned a lot from it.

Very happy to host @MatthiasRottner in Zurich today @uzh_df @SFI_CH. Matthias presented "Estimating Nonlinear Heterogeneous Agent Models with Neural Networks" (with @HannoKase & @LeonardoMelosi). Deep Learning + HANK + Estimation!

@YuchengYang1993 @uzh_df @SFI_CH @HannoKase @LeonardoMelosi Thank you very much for hosting me! It was a great day.

The Bundesbank's research centre has resources to hire one or more visiting macro researchers, on contracts from 6 months to 2 years. Ideal for a sabbatical. We would also consider new PhDs who want to spend some time with us before starting a job elsewhere. DM me if interested.

After the recent Nobel prizes, many economists are wondering how recent advances in deep learning may affect economics galonuno.com/uploads/1/3/4/…

This article develops new geopolitical risk indicators based on local newspaper coverage and study the transmission of geopolitical risk shocks using Russia as a case study. Shocks identified using local news sources have significant adverse effects on the Russian economy. 2/2