image

Mohammad Ebrahimi

Finance
Cohort 2023
HEC Montréal


Program of study

Finance

University

HEC Montréal

Academic degree

Master's


Academic background

I hold an M.Sc. in Finance from HEC Montréal (2022-2024), where I completed a thesis on predicting bubbles in the real estate market using sentiment analysis. My academic background also includes an M.Sc. and a B.Sc. in Civil Engineering from Sharif University of Technology. During my studies, I have been involved in both research and teaching. At HEC Montréal, I worked as a research assistant on projects related to financial econometrics and machine learning, and I served as a teaching assistant for Financial Econometrics. Additionally, I was a teaching assistant for Game Theory at Sharif University of Technology.

About me

With an engineering background and diverse research experience, I have developed strong technical skills in Python, machine learning, and econometrics, along with a solid foundation in quantitative analysis. I am particularly interested in exploring the applications of AI in the finance industry and am eager to pursue this passion further.

Outputs


Predicting Housing Prices in Canada Using News Sentiments, Bubble Indicators, and Google Trends

In the thesis, we investigated the predictive power of news sentiment, bubble indicators, and Google Trends data for forecasting real estate prices. To quantify sentiment, we utilized large language models (LLMs) to extract a sentiment index from news articles. We also incorporated bubble indicators, specifically the SADF and BSADF statistics from the PSY test, along with A and S indicators from the WHL test. Additionally, we introduced a novel Housing Search Index based on Google Trends data. Using forecast-encompassing tests on Autoregressive models with exogenous inputs (ARX), we assessed the individual forecasting performance of each predictor. My findings indicate that, in certain cases, news sentiment and bubble indicators enhance forecast accuracy. Notably, Google Trends-based indices outperformed other predictors at the one-month forecasting horizon. This project led to a Python package called GTBpy: https://github.com/Mohammad-ebr/GTBpy. Thesis article can be found at: https://hecmontreal.on.worldcat.org/oclc/1458653247


Internships Details

Currently looking for an internship

Prefered Starting Date: 01/09/2024

Prefered Theme: Quantitative Finance


image image image image image image image image