Jiahe Lin

Ph.D. in Statistics

I am a Quantitative Researcher at an asset management firm, developing deep learning models that leverage core statistical backbones to capture complex dynamics from market data.

Previously, I was a member of the Machine Learning Research team at Morgan Stanley, where I designed and deployed deep learning models for forecasting and decision-support systems across equity and fixed income markets. I also contributed to academic research in the intersection of deep learning and statistics, with papers published in leading ML and statistics venues. Prior to this, I was a Quantitative Strategist at Goldman Sachs, building statistical and econometric models.

My academic training was in statistics, and my graduate research focused on high-dimensional time series and graphical models, as well as optimization algorithms.

Broadly, I am interested in:

  • Interpretable deep learning models for complex dynamics (e.g., state-conditioned and contextual architectures)
  • Generative models as flexible frameworks for probabilistic simulation, latent variable modeling, and structured inference (e.g., VAEs, diffusion, and flow models)
  • Modern LLM system mechanics – such as retrieval systems (RAG), RL, and pre/post-training pipelines – and their adaptation to time series
  • Causal discovery and inference in observational data