Nicolas Harvie

Nicolas Harvie

PhD Candidate in Finance
Rotman School of Management
University of Toronto

Rotman School of Management
105 St George St
Toronto (Ontario)
Canada ON M5S 3E6

Code

My publicly available code is distributed through GitHub.

Select Projects

  • Financial Risk Management Project Python, R & Matlab The main objectives are to forecast risk measures for benchmark portfolios, and assess the risk of adding new strategies to existing portfolios, both statically and dynamically. Additionally, we aim to improve the tool incrementally by incorporating new advanced methodologies such as multivariate volatility forecasting, copula models, extreme value theory and machine learning.

  • Financial Econometrics Labs Python Labs that were developed as a showcase for adequate practice to help students complete their assignments in MATH60210. Topics covered include asset pricing models, financial time series prediction, model estimation, risk and volatility modeling, and event studies.

  • Anomaly Research Python We know that anomalies to the efficient market lose significance over time, and markedly so after publication. Our first objective is to assess and demonstrate the existence of post-publication decay and loss of statistical of market anomalies over time. In a second stage, we use information about the alpha decay of clear predictors of excess returns to gain information (correlation, threshold correlation, etc) about the select anomalies who seem to hold in signficance through time

  • Macro Factor Timing Python Based on the insights from “Macro Trends and Factor Timing” (Favero & al, 2022), we seek to create a factor timing equity trading strategy based on the cointegration logic between factor returns and macroeconomic variables. First, we replicate the article in the exact same conditions as is presented. Then, we modify the framework to include the strongest predictors of excess returns.

  • Cryptocurrency Momentum Strategy In Collaboration with Hugo Couture Python, R This research project presents a crypto-based trading strategy that integrates price momentum, size, and investor attention to create a robust and efficient investment approach. The foundation of the strategy is derived from the research conducted by Yang (2019), Liu (2021), and Liu (2022), which collectively support the inclusion of momentum, size and attention in the trading framework.

  • Mixed Momentum Strategy Python Based on limits to arbitrage arguments and conclusions derived from our experiments in Anomaly Research, we devise a stock trading strategy effectively based on the interaction of three signals (i) uncertainty, as proxied by firm age (Zhang, 2006) (ii) individual stock return momentum (Jegadeesh, 1993) (iii) announcement return, proxied by standardized change in earnings or abnormal returns (Zhang, 2006).

  • FF Decimals Randomization Python Using Chen and Zimmerman (2020) panel of clear predictors, we determine the impact of rounding errors induced by the decimals approximation implied in the construction of the Fama-French factors, on two types of asset pricing tests (Fama & French, 1992) (Fama & Macbeth, 1973). Conclusions to be determined…

  • Corporate Finance in Stata Stata Objective is to replicate the same important empirical corporate finance papers, but in Stata to get a working knowledge of the language.