I'm a PhD candidate at the Vienna University for Economics and Business. My research interests lie in applied econometrics, environmental economics, and development economics.

You can check out my CV , contact me via mail or phone , or find out more about my research , software , and teaching below.

Research

I'm interested in practical econometric methods for investigating spillover effects, networks, and uncertainties, as well as issues related to deforestation and development. My recent work includes ...

• Kuschnig, N., Zens, G., and Crespo Cuaresma, J. (2022). Hidden in plain sight: influential sets in linear regression. PDF: .

  The sensitivity of econometric results is central to their credibility. In this paper, we investigate the sensitivity of regression-based inference to influential sets of observations and show how to reliably identify and interpret them. We explore three algorithmic approaches to analyze influential sets, and assess the sensitivity of a number of earlier studies in the field of development economics to them. Many results hinge on small influential sets, and inspecting them can provide crucial insights. The analysis of influential sets may reveal omitted variable bias, unobserved heterogeneity, lacking external validity, and informs about technical limitations of the methodological approach used.

• Kuschnig, N. (2023). Shrinkage in space — spillovers in a Bayesian hierarchical model. PDF: .

  In this paper, I present a modelling approach to jointly investigate connectivity between observations and its consequences — spillover effects. The approach is fully Bayesian, and uses hierarchical shrinkage priors to flexibly provide regularization where needed and let the data speak where it is possible. I make the prior information that is embodied in the restrictive assumptions of previous spatial models explicit, and loosen them by estimating connectivity parameters. For effective estimation, I develop efficient sampling procedures and a Gaussian process approximation to evaluate Jacobian determinants.

• Kuschnig, N., Vashold, L., Soterroni, A., and Obersteiner, M. (2023). Eroding resilience of deforestation interventions — evidence from Brazil’s lost decade. R&R at Environmental Research Letters, PDF: .
• Kuschnig, N., Vashold, L. (2022): The economic impacts of malaria: past, present, future. Planetary Health, PDF: .

Publications

• Giljum, S., Wegner Maus, V., Kuschnig, N., Luckeneder, S., Tost, M., Sonter, L., and Bebbington, A. (2022). A pantropical assessment of deforestation caused by industrial mining. Proceedings of the National Academy of Sciences, 119(38):e2118273119, DOI: 10.1073/pnas.2118273119.
• Kuschnig, N. (2022). Bayesian spatial econometrics: a software architecture. Journal of Spatial Econometrics, 3(1):6–25, DOI: 10.1007/s43071-022-00023-w.
• Kuschnig, N. and Vashold, L. (2021). BVAR: Bayesian vector autoregressions with hierarchical prior selection in R. Journal of Statistical Software, 100(14):1–27, DOI: 10.18637/jss.v100.i14.
• Kuschnig, N., Crespo Cuaresma, J., Krisztin, T., and Giljum, S. (2021). Spillover effects in agriculture drive deforestation in Mato Grosso, Brazil. Scientific Reports, 11(1):1–9, DOI: 10.1038/s41598-021-00861-y.
• Kuschnig, N. (2021). Inadequate methods undermine a study of malaria, deforestation and trade. Nature Communications, 12(1):1–3, DOI: 10.1038/s41467-021-22514-4.
• Bruckner, M., Wood, R., Moran, D., Kuschnig, N., Wieland, H., Maus, V., and Börner, J. (2019). FABIO — the construction of the food and agriculture biomass input-output model. Environmental Science & Technology, 53(19):11302–11312, DOI: 10.1021/acs.est.9b03554.

Software

I'm passionate about free and open source software, and have written packages for Bayesian modelling, efficient computation, and data handling in R.

• Kuschnig, N. and Vashold, L. (2023). BVAR: hierarchical Bayesian vector autoregression. R package available on CRAN, paper at DOI: 10.18637/jss.v100.i14.
• Kuschnig, N. (2022). bsreg: Bayesian spatial regression models. R package on CRAN, paper at DOI: 10.1007/s43071-022-00023-w.
• Kuschnig, N. (2023). influence: Sensitivity to influential sets. R package on GitHub.
• Kuschnig, N. (2020). sanic: solving Ax = b nimbly in C++. R package on CRAN.
• Vashold, L. and Kuschnig, N. (2020). BVARverse: tidy Bayesian vector autoregression. R package on CRAN.

Teaching

I have taught courses at the Master's and Bachelor's level at WU, served as a teaching assistant at CEU, and organised two reading groups with various guests at WU.

• Econometrics 2 (WU Bachelor)

  Econometric methods, focused on causal inference, and supplemented with assignments that involve applied coding and prediction tasks. Topics include an introduction to statistical learning, causality, experiments, directed acyclic graphs, instrumental variables, non-linear models, maximum likelihood estimation, regularization, and methods for causal inference. Slides are available here .

• Bayesian Macroeconometrics (WU Master)
• Applied Econometrics (WU Bachelor)
• Econometrics 2 (CEU Master, TA)

Reading group

I organised and headed two reading groups at WU. The goal was for participants to discuss a range of economics- and research-related issues amongst themselves, and with experienced guests.

• Topics and guests included ...

  • evidence-based consulting with Gabriel Felbermayr,
  • the economics of automation with Klaus Prettner,
  • the Austrian micro data center with Harald Oberhofer,
  • uncertainty and growth with Jesús Crespo Cuaresma,
  • economics and politics of migration with Peter Brummund,
  • the Austrian national budget with Markus Marterbauer,
  • economic policy consulting with Christoph Badelt,
  • monetary policy in practice with Ewald Nowotny,
  • land use policy in Austria with Barbara Birli,
  • labour market experiments with Lukas Lehner & Anna Schwarz,
  • corporations and economic policy with Wilfried Altzinger,
  • new approaches to monetary policy with Lea Steininger,
  • green growth with Colleen Schneider & Thomas Neier,
  • incarceration and the racial divide with Fabian Siuda.