Rio de Janeiro, Brazil
B.S., Brazilian School of Economics and Finance; M.S., Graduate School of Economics
"I am a Princeton engineer because I joyfully follow curiosity wherever she leads and make sure to follow up with execution."
Laura’s research is in the area of modeling high-frequency trading. Her goal is to understand the behavior of agents in financial markets. In her most recent project, she studies optimal execution. Optimal execution refers to an investment firm’s duty to execute orders on behalf of their clients, sometimes for many clients and millions of dollars all at once. She applies a machine learning model to market statistics to optimize and estimate best practices for traders. The aim of her research is to see how traders can behave optimally in the market with the least amount of impact to the price process. If traders do not behave optimally, they can negatively affect their own profit and loss, as well as their clients’ investments. At high volumes, whole market ecosystems can be damaged, causing instability or crashes.
At the end of her first project, she successfully created an accurate model for optimal trading behavior. In her second project, she is seeking to provide model explainability and translation. This can be especially useful for regulators who set market guidelines, ensuring the safety of trading for all market players. In order to do this, she built a theoretical framework to authorize neural network explainability. This means that she studied how and why the neural networks came up with an optimal solution, which helps regulators verify the model’s dependability. Her approach of using neural networks is a significant improvement on the previous state-of-the-art, which relied on partial differential equations, mathematical models that rely on many assumptions that were not always indicative of behavior. The neural network approach offers a more flexible model that is based on actual behavior.
Laura worked at an investment bank prior to applying to Princeton. With work experience behind her, she wanted to make sure that her time in graduate school was doing what she loved: research. She sought out a great research institution and found Princeton’s ORFE program to be her top pick. After getting accepted, on her campus visit, she felt a good energy in Princeton, which solidified her drive to attend.
When she first got to campus, immediately after setting up her apartment and necessities, she signed up for “every Princeton event.” She found the campus to be inviting with so many activities and ways to get involved that she was inspired to meet “every soul at this university,” which she admits was “obviously impossible.”
"There’s such an opportunity to meet people from other fields like plasma physics or history or Scandinavian languages,” she says, noting that in her first week she thought, “wow, this is like a universe for curious persons. It’s like a dream.”
Since being at Princeton, she has found her advisor René Carmona to be so supportive, “an out-of-the-ordinary good advisor.” She says that he “goes above and beyond to help his students thrive." Being personable and open to candid discussions was very important to Laura because she is from Brazil, which she describes as a very warm and inviting culture. Carmona’s group was a perfect fit, she says.
Early Interest in STEM
She became interested in the stock market as early as age 15. At that time, she asked her father to buy her stock so that she could trade on his account. He obliged, and that began her fascination with trading.
Before that, as early as 5 years old, her father was teaching her about compound interest and the power of saving and investing her allowance.
These early encounters led her to pursue bachelor and master's degrees in economics before joining Princeton’s doctoral program in ORFE.
School of Engineering and Applied Science Award for Excellence, Princeton University, 2020.
Undergraduate Teaching Award, Princeton University, 2019.
Gordon Wu Fellowship, Princeton, University, 2016.
L. Leal, C. Almeida. An SDF Approach to Hedge Funds’ Tail Risk: Evidence from Brazilian Funds. Brazilian Review of Econometrics, 37(1):61-88, 2017.
L. Leal, M. Laurière, C.A., Lehalle. Learning a Functional Control for High-Frequency Finance. arXiv preprint arXiv:2006:09611 (2020). (Under review)