Research

The distribution of individual causal effects is not identifiable due to the fundamental problem of causal inference. I work on identifiability assumptions and characterize the difference between (conditional) average effects and individual effects.

Related papers:

• Post RAJ, et al. Flexible Machine Learning Estimation of Conditional Average Treatment Effects: A Blessing and a Curse. Epidemiology, 2024;35(1);32. 

• Post RAJ & van den Heuvel ER. Beyond Conditional Averages: Estimating The Individual Causal Effect Distribution. arXiv preprint, 2024.  

• Post RAJ et al. Individual causal effects from observational longitudinal studies with time-varying exposures. arXiv preprint, 2023. 

Many epidemiological studies focus on time-to-event outcomes.  Traditional measures of effect are based on the hazard rate and have a complicated causal interpretation. Formalizing this interpretation is crucial for accurate causal inference.

Related papers:

• Post RAJ, et al. The built-in selection bias of hazard ratios formalized. Lifetime Data Analysis, 2024; 30;404.

• Post RAJ, et al. Bias of the additive hazard model in the presence of causal effect heterogeneity. Lifetime Data Analysis, 2024; 30;383. 

• Brathovde M, et al. Formalizing the causal interpretation in accelerated failure time models with unmeasured heterogeneity. arXiv preprint, 2024. 

I have been involved in several interdisciplinary collaborations between the mathematics and chemistry departments at the Institute for Complex Molecular Systems (ICMS) at the Eindhoven University of Technology.

Related papers:

• Post RAJ, van der Zwaag D, et al. A stochastic view on surface inhomogeneity of nanoparticles. Nat Commun. 2019;10(1):1663. 

• Post RAJ, Song S, Mason AF, et al. “Engineering transient dynamics of artificial cells by stochastic distribution of enzymes.” Nat Commun. 20201;12(1):6897 

• Post RAJ, et al. Interevent-time distribution and aftershock frequency in non-stationary induced seismicity. Sci Rep. 2021;11;3540

After my move to the Erasmus MC in October 2024, I started collaborations with medical doctors and epidemiologists to apply causal inference methods in practice. The results of those collaborations will appear in the future.