Mutual funds in North and South America: Relationship between news and Mutual funds Returns

• Název v anglickém jazyce: Mutual funds in North and South America: Relationship between news and Mutual funds Returns
• Klíčová slova anglicky: Asset Pricing, Mutual Funds, Portfolio Management, Financial Markets, Risk Management
• Akademický rok vypsání: 2022/2023
• Typ práce: diplomová práce
• Jazyk práce: angličtina
• Ústav: CERGE (23-CERGE)
• Vedoucí / školitel: PhDr. Mgr. Ctirad Slavík, Ph.D.
• Řešitel: skrytý - zadáno a potvrzeno stud. odd.
• Datum přihlášení: 04.04.2023
• Datum zadání: 04.04.2023
• Datum potvrzení stud. oddělením: 04.04.2023

Seznam odborné literatury

French, C. W. (n.d.). The Treynor Capital Asset Pricing Model. Retrieved March 30, 2023, from https://papers.ssrn.com/sol3/papers.cfm?abstract_id=447580

Fraiberger, S. P., Lee, D., Puy, D., & Ranciere, R. (2021). Media sentiment and international asset prices. Journal of International Economics, 133, 103526. https://doi.org/10.1016/j.jinteco.2021.103526

Stock Prediction by Integrating Sentiment Scores of Financial News and MLP-Regressor: A Machine Learning Approach. (n.d.). Procedia Computer Science, 218, 1067–1078. https://doi.org/10.1016/j.procs.2023.01.086

Předběžná náplň práce v anglickém jazyce

In my thesis, I want to investigate the connection between Google Trends data and media news with the North and South American mutual fund returns.
For that matter, I have prepared the following questions:
“What is the relationship between news and mutual fund returns in North and South America? Can sentiment analysis of news articles help to assess mutual fund returns?”

Originality: The concept is unique in that it focuses on mutual fund returns rather than stock market returns, even though there is current research on the relationship between news sentiment, Google Trends data, and stock market returns. The thesis will also offer insights into possible variations in these factors across North and South America, which has not been widely studied, considering the differences in each region between North and South America.

Importance: The thesis topic has interesting implications for investors and asset managers interested in creating more precise forecasting models for mutual fund returns. Investors may be better able to predict changes in mutual fund performance by combining news sentiment and Google Trends data into asset pricing models.

Theoretical Context: There are several asset pricing models. The most common ones are the CAPM (Capital Asset Pricing Model), the Fama-French Model ( Which is an extension of the CAPM model that adds three, five, and six factors), and APT ( Arbitrage Pricing Theory). The standard Fama-French 3-factor model and ATP Model are shown below
Fama-French Model: Ri - Rf = αi + βi1(Rm - Rf) + βi2SMB + βi3HML + εi (models like the Fama-French, focus on the market characteristics “market risk, value, size,
momentum”) APT Model: Ri = Rf + β1F1 + β2F2 + β3F3 + εi (APT model assumes a variety of factors e.g. can explain return macroeconomic factors therefore F1, F2, F3 can be any factor like GDP growth, inflation, etc.) So, the objective would be to take one model and modify it so it can add news sentiments and/or Google trend data.

Methodology: A comprehensive literature review will serve as the foundational step for this thesis, providing insights into existing research and methodologies in the field. Building on the findings from the literature review, a modified version of asset pricing models will be employed. This adaptation allows for incorporating news sentiments and Google Trends data, which are novel components of this study. Given the limitations in data availability for Latin America, the research will focus on utilizing data from the largest and most comprehensive stock exchanges in the region, including exchanges in Mexico, Brazil, Chile, among others. Additionally, advancements in Natural Language Processing (NLP) technology will facilitate the collection and analysis of news data, enhancing the depth and scope of the research. The Analysis is expected to include Sentiment Analysis, Regression Analysis, and Statistical Testing. The research may differentiate between North and South America to account for potential variations. This differentiation may involve separate models or the inclusion of region-specific variables and interaction terms. Analyzing regional differences will provide insights into how news sentiment and Google Trends data influence mutual fund returns in distinct geographical contexts. The study will conclude by summarizing the findings and their practical implications for investors and asset managers. Specifically, it will highlight how incorporating news sentiment and Google Trends data can enhance the accuracy of mutual fund return forecasting models. Acknowledging the inherent limitations of the study, such as data availability constraints, the research will suggest avenues for future exploration. Potential future research areas may include exploring additional factors or refinements to the methodology.

Outline:

Introduction
Theoretical foundations
Methodological Approach
Analysis and Results
Conclusion and Suggestions
References