Quant Research Workflow and Alternative Data for Strategy Development
Summary
This trip report summarizes ideas from a quant meetup and trading conference, with its most concrete trading content focused on strategy research. A talk described applying vertical improvement to an existing approach and horizontal exploration of new markets or data. For post earnings announcement drift, the speaker combined analyst earnings surprise data with news sentiment data, reporting a Sharpe ratio of 1.36 over the tested period. The report gives no details on the test design, costs, universe, or robustness, so that figure alone cannot establish live performance.
The author’s own presentation outlines a professional research workflow: form an investment thesis, acquire and clean data, define a universe, develop signals, combine models, construct and manage a portfolio, and execute trades. Other conference discussions touched on discretionary and quantitative methods, computing, and programming language tradeoffs. These are reported as discussion topics rather than validated research findings. The central practical lesson is to improve the full research process and investigate less commonly used data, while carefully evaluating whether apparent alpha survives rigorous testing.
Key ideas
- Strategy research can explore new datasets as well as refine established signals.
- The reported earnings drift example combined analyst surprise data with news sentiment and achieved a Sharpe ratio of 1.36 in its stated test.
- The report does not provide enough testing detail to assess the robustness or tradability of that result.
- A structured quant workflow moves from investment thesis and data preparation through portfolio construction, risk management, and execution.
- Programming tools, computing methods, and human judgment were discussed as parts of the broader research environment.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.