Jiajing Sun

Changing units or reordering coefficients should give the same inference about the underlying economic quantities. Yongmiao Hong, Zhuo Lin, Oliver Linton, Whitney K. Newey and Jiajing Sun develop joint confidence regions for dependent time series without estimating a long-run covariance matrix or choosing a bandwidth. Their application studies US fiscal multipliers. The paper appears in the University of Cambridge Janeway Institute working paper series (No. 2637).

Jiajing Sun co-authors and maintains aersn, the R and Python implementations of the method.

Cambridge working paper · R package (CRAN) · Python package (PyPI) · Method and confidence-region animations

My methodological work centres on time-series econometrics, including the Adjusted-Range Self-Normalisation framework for robust inference on long-run variances under strong dependence and heavy tails — alongside functional time series and self-normalised monitoring.

On the applied side, my work spans financial risk, digital finance, environmental management, and ecological economics. This includes regulatory early-warning research for derivatives markets, text- and machine-learning-based evaluation of financial-technology competitiveness, and virtual water in global supply chains.

I present regularly at international meetings in econometrics and statistics — among them the World, European, Asian, North American, and Australian meetings of the Econometric Society, SoFiE, and CMStatistics — and at invited university seminars. Slides are shared where I can.

I teach postgraduate courses in financial econometrics, advanced time series, regression, and quantitative finance, and write textbooks that connect statistical theory with empirical application and reproducible code in R and Python. Companion websites bring together chapter guides, lecture slides, videos, and coding resources for teaching and independent study.