Shaikh Tanvir Hossain

Doctoral Candidate
TU Dortmund

Waterfall in forest

Photo by Blake Verdoorn on Unsplash

Dissertation

Parametric and Nonparametric Learners for Adaptive Experiments Under Interference

Supervisor: Professor Carsten Jentsch

My doctoral research focuses on adaptive experimentation or multi-armed bandits. Broadly, I study how classical multi-armed bandit frameworks must be adapted when the outcome for one unit depends not only on the treatment it receives, but also on the treatments assigned to other units with whom it interacts. This phenomenon is known as interference. My work develops a unified theoretical framework for bandit problems under interference, proposes a family of parametric and nonparametric algorithms tailored to different interference structures, and connects the classical bandit setting to this more general interference setting.

A working paper based on this research is currently being finalized for submission. Details will be shared here shortly.

Working Papers

Bias Correction, Bootstrap, and Kernel Estimators

We propose weighted bootstrap confidence intervals for the bias-corrected local linear estimator, building on the bias-correction approach of Calonico, Cattaneo, and Farrell (2018) and Cheng and Chen (2019), and study pointwise performance.

Autoregressive Dynamic Network Modeling with Serial and Cross-Sectional Dependence

Joint work with Professor Carsten Jentsch; an extension of my master’s thesis.

Presentations

  • German Statistical Week, 2021 (online), University of Kiel. Nearest Neighbor Matching: Does the M-out-of-N Bootstrap Work When the Naive Bootstrap Fails?

  • German Probability and Statistics Days (GPSD), 2018, University of Freiburg. Modeling and Prediction of Dynamic Networks Using Binary Autoregressive Time Series Processes.

Workshops and Conferences

  • Bielefeld-Dortmund Summer School 2017 on Modern Topics in Time Series Analysis, TU Dortmund.
  • Workshop on Causality (DFG Research Unit 1735, “Structural Inference in Statistics”), Uder. Lectures by Professor Jonas Peters and Professor Vanessa Didelez.
  • Workshop on Mathematical Statistics (“StatMathAppli 2019”), Fréjus. Lectures by Professor Matthieu Lerasle and Professor Peter Bühlmann.