Wolfram Computation Meets Knowledge

Wolfram Summer School


Nam Tran-Hoang

Science and Technology

Class of 2019


Nam likes math and science. As a self-respecting chemistry and mathematics graduate from Ohio Wesleyan University, he dreamed of holding a tenured position in academia, teaching younger generations about his beautiful subject(s)—until reality woke him up and pushed him to make an immediate use of his skills. He is currently studying finance at University of Miami. In his free time, Nam enjoys reading about the business world and the many “creative” schemes people devise to make money, such as “mortgage-backed collateralized debt obligation” or “high frequency trading” (he strongly recommends Michael Lewis’s books). He learns and solves math/logic puzzles for fun also (he usually sources the puzzles/topics from Brilliant, 3blue1brown). To keep himself sane, Nam swims regularly. He also listens to classical music (Bach, Paganini) and absurd pop song romance.

Computational Essay: Plasma Drug Concentration

Project: Investment Strategies: Functional Weighted Choice Method


A quantitative model was derived to analyze common collections of investment tactics. The model was designed with flexibility: it can choose and weight the stocks into a portfolio of any size, based on any analytic expression of the stocks’ financial characteristics. Using the model, several common investment strategies were investigated: chosen and weighted by price; chosen by price, equally weighted; and chosen by volume, weighted by market capitalization. Consistent with the capital asset pricing model, the model shows that a well-diversified portfolio exhibits less volatility and less return. The model also indicates that the frequency of adjustment does not exert a strong effect on the strategy’s performance.

Summary of Results

The project constructs a computation model to analyze investment strategies. The model can choose and weight stocks into a portfolio of any size, based on any analytic expression of the stocks’ financial characteristics. The model also allows different specification of adjustment frequency: daily, every other day, biweekly, etc.

Future Work

Based on this computational model, the optimal investment tactics can be determined for a finite set of strategies, for example value-based, growth-based, simple arithmetic expressions, etc. Furthermore, the financial characteristics of distinct strategy sets can be investigated to elucidate their common properties and their similarities and differences.