New opportunity

Research Fellow (Econometrics and Machine Learning)

Nanyang Technological University · Singapore

About this role

The Economics Programme in NTU’s School of Social Sciences is seeking a Research Fellow to contribute to research on reliable instrumental variable (IV) estimation and inference methods, as well as machine learning algorithms for high-dimensional data and weak instruments. Key Responsibilities: Develop econometric theory for weak-identification-robust tests, bootstrap methods, estimator combinations, and leave-cluster-out bias correction. Conduct simulations and empirical analyses on topics including import competition, education, and pretrial decisions using IV regressions. Collaborate with the project team to produce reproducible code, research papers, technical appendices, and presentations. Job Requirements: A PhD in Economics, Econometrics, Statistics, or a closely related field is required by the appointment start date. Candidates must possess strong skills in econometric theory, probability, statistics, IV methods, causal inference, and machine learning. Proficiency in R, Python, MATLAB, or Julia is essential. Strong communication and collaboration skills are required. Desirable Qualifications: Specialist experience in weak identification, bootstrap methods, or cluster-robust inference is considered an asset. Prior postdoctoral experience is also desirable. The position welcomes entry-level PhD candidates who meet the qualification requirements by the start date.

For full details about this role, click “Apply on company site” to view the employer’s original posting.

Skills for this role

EconometricsMachine LearningInstrumental Variable (IV) estimationCausal inferenceRPythonMATLABJuliaStatistical analysisSimulation modelingBootstrap methodsCluster-robust inference

Checking your profile…

Preparation suggestions, not the employer’s actual interview questions. Always check the original posting for current requirements.

Understand the role

What matters most

Development of Econometric Theory
The core of the role involves advancing theoretical frameworks for weak-identification-robust tests, bootstrap methods, and bias correction. You must demonstrate the ability to translate complex statistical concepts into rigorous, publishable econometric theory.
Integration of Machine Learning with IV Estimation
The role requires applying machine learning algorithms to address high-dimensional data and weak instruments. You should be prepared to discuss how ML techniques can be effectively combined with traditional causal inference and IV regression methods.
Reproducible Research and Technical Documentation
A critical duty is producing reproducible code and technical appendices. You must be proficient in R, Python, MATLAB, or Julia and capable of maintaining high standards for code transparency and documentation to support collaborative research papers.

Likely challenges to think through

Balancing Theoretical Rigor with Empirical Application
The role requires both developing new theory and applying it to empirical topics like import competition or pretrial decisions. Prepare by articulating how you maintain theoretical integrity while ensuring your models remain practical and robust for real-world data analysis.
Managing Computational Complexity in Simulations
Implementing simulations for high-dimensional data and weak instruments can be computationally intensive. Be ready to discuss your strategies for optimizing code performance and ensuring the stability of your estimation methods under challenging data conditions.

Practice your answers

Technical practice

Addressing Weak Instrumental Variables

What to coverExplain the challenges posed by weak instruments in IV estimation and discuss specific robust inference methods you would implement to mitigate these issues. Focus on the trade-offs between size control and power in weak-identification-robust tests.

Machine Learning for High-Dimensional Data

What to coverDescribe how you would integrate machine learning algorithms with traditional econometric models when dealing with high-dimensional instruments or controls. Discuss the regularization techniques you would use to prevent overfitting while maintaining causal interpretability.

Cluster-Robust Inference and Bias Correction

What to coverDiscuss the implications of clustered data on standard error estimation and the theoretical basis for leave-cluster-out bias correction. Explain when and why you would prioritize these methods over standard cluster-robust variance estimators.

Bootstrap Methods in Econometrics

What to coverCompare and contrast different bootstrap resampling techniques for econometric models. Explain how you would validate the reliability of bootstrap methods when the underlying data structure involves complex dependencies or weak identification.

Reproducible Research Practices

What to coverDetail your workflow for ensuring research reproducibility when collaborating on complex econometric projects. Address how you manage version control, documentation of technical appendices, and the integration of code across R, Python, or Julia.

Causal Inference in Empirical Analysis

What to coverGiven a scenario involving import competition or pretrial decisions, outline your approach to identifying a valid causal effect. Explain how you would test the exclusion restriction and address potential endogeneity in your IV regression model.

Simulation Modeling for Estimator Evaluation

What to coverDescribe your process for designing Monte Carlo simulations to evaluate the performance of new econometric estimators. Explain how you determine the parameter space and sample sizes necessary to demonstrate the robustness of your proposed methods.

Estimator Combinations

What to coverExplain the theoretical advantages of combining multiple estimators versus selecting a single 'best' estimator. Discuss how you would handle the weighting of these estimators to minimize mean squared error in finite samples.

Behavioural practice

Collaborative Research Challenges

What to coverDescribe a situation where you had to collaborate with a team to produce a research paper or technical report. Focus on how you managed differing technical opinions or integrated feedback to reach a final, high-quality output.

Communicating Complex Theory

What to coverProvide an example of a time you had to explain a complex econometric or machine learning concept to a non-specialist or a broader research audience. How did you adapt your communication style to ensure clarity without sacrificing technical accuracy?

Handling Research Deadlines and Ambiguity

What to coverDescribe a time when your research hit a significant technical roadblock or unexpected result. Explain the steps you took to troubleshoot the issue and how you adjusted your project plan to meet your research objectives.

Managing Multiple Research Streams

What to coverDiscuss a time you balanced multiple research tasks or projects simultaneously. Explain your prioritization strategy and how you ensured that the quality of your empirical analysis and code remained consistent across all work.

Sharpen your knowledge

Topics to prepare

Weak Identification and IV Robust Inference
The role focuses on developing theory for weak-identification-robust tests. Review literature on weak instrument asymptotics, specifically focusing on Anderson-Rubin tests and conditional likelihood ratio tests. Exercise: Derive the asymptotic properties of a test statistic under weak identification and compare its size control against standard Wald-type tests in a simulation.
Cluster-Robust Inference and Bias Correction
The position requires expertise in leave-cluster-out bias correction and cluster-robust inference. Review the literature on clustered standard errors and the impact of small cluster numbers. Exercise: Implement a leave-cluster-out cross-validation procedure in a simulated dataset to assess its performance in reducing bias compared to standard cluster-robust variance estimators.
Machine Learning for High-Dimensional Causal Inference
The project involves applying machine learning to high-dimensional instruments or controls. Review double machine learning (DML) frameworks for causal parameter estimation. Exercise: Apply a Lasso or Random Forest-based regularization technique to a high-dimensional IV regression model and evaluate the consistency of the structural parameter estimate.

Skills to revise

Simulation Modeling in R or Python
The role requires implementing simulations for econometric theory. Exercise: Create a reproducible script in R or Python that generates synthetic data with weak instruments and clustered errors, then compare the performance of different IV estimators (e.g., 2SLS, LIML, and JIVE) across varying sample sizes.
Reproducible Research Workflow
The team emphasizes the production of reproducible code and technical appendices. Exercise: Organize a past research project into a clean, modular repository structure using version control, ensuring that all empirical results and figures are generated automatically from raw data via a master script.

Your preparation checklist

Prepare a technical portfolio of reproducible research
Compile examples of your previous work that demonstrate proficiency in R, Python, MATLAB, or Julia. Ensure your code is well-documented and includes technical appendices that show your ability to implement complex econometric simulations or machine learning algorithms.
Synthesize your experience with weak identification and inference
Be prepared to discuss your specific research contributions regarding weak-identification-robust tests, bootstrap methods, or cluster-robust inference. Frame your past projects to highlight how you addressed bias correction or high-dimensional data challenges.
Develop a research narrative for the project topics
Review your familiarity with the specific empirical domains mentioned in the job description, such as import competition, education, or pretrial decisions. Prepare to explain how your econometric toolkit can be applied to analyze these specific areas using IV regressions.
Formulate questions regarding team collaboration and research output
Prepare questions for the interviewers about the current project team's workflow, how research papers are co-authored, and the expected balance between developing new econometric theory and conducting applied empirical analysis.

Availability last checked: 18 Oct 2026. Status can change after a check; confirm on the employer’s website. Report an issue · Our checking policy

KEEP EXPLORING

Similar AI jobs

Related skills and specializations in Singapore. Matched to this role, not your profile.

Explore more jobs

Finding similar opportunities…