Research Projects
Money and Banking
Prediction of the 5/1-Year Adjustable-Rate Mortgage Average in the United States using the Federal Reserve Beige Book with Natural Language Processing and Time series model
I conducted a research project employing data science, time series and data mining tools to forecast the 5/1-Year Adjustable-Rate Mortgage Average in the United States using the Federal Reserve Beige Book with Natural Language Processing and Time series models. The qualitative aspect of the Beige Book made this project very inspiring, but the results were worthwhile; I used some sophisticated Data Mining tools to extract some key information from this document which is the only document accessible to the public.
The most crucial finding of this paper is that the polarity of the Beige Book does affect, to a certain extent, the adjustable-rate mortgage. Using multiple methods allowed us to confirm that relationship. While the simple linear regression hints at that relationship, the time-series models like the VAR and ARIMAX demonstrate that relationship.
Time Series
Time series modeling and analysis of the Chicago Board Options Exchange (CBOE) Volatility Index (VIX Index)
I conducted a group research project related to Time series modeling and analysis of the Chicago Board Options Exchange (CBOE) Volatility Index (VIX Index) to generate forecasts as close to the actual forecasting data as possible. With the erratic nature of this index, it can be challenging to build the perfect model that can capture the movement of this index. Fortunately, with my previously acquired knowledge of financial and stock markets, I successfully conducted this project.
Data Mining
Exploration and analysis Autistic Spectrum Disorder Screening Data for Adults with Gradient Descent, Classification using TensorFlow, and Neural Network
I explored, researched, and analyzed Autistic Spectrum Disorder Screening Data for Adults with Gradient Descent, Classification using TensorFlow, and Neural Networks. Learning new concepts, tools and approaches was exigent and demanding at first. However, with my passion for data science and data analysis, I enjoyed undertaking this project and improving my data science and programming skills. In fact, I found some astonishing results.
Originally from a developing country where the science of data is not developed, I enjoyed every minute of this class, and the projects and case studies increased my passion for that field, that I consider complementary to most of the other fields.
Labor Economics
Factors affecting the average hourly wages in the U.S. using quantile and linear models
I conducted a project to empirically validate factors affecting the average hourly wages in the U.S using quantile and linear regressions. This type of project can be testing, as in social sciences, we often have some apprehensions related to specific subjects, and conjectures about the research results may be misleading. However, what I appreciate in this type of project is that the data sometimes goes against the primary apprehension we can have about the causes of a phenomenon. This project changes my previous point of view about this subject.
​Game Theory
Research use of sequential game models to identify what Taiwan can learn from Russia's invasion of Ukraine
I conducted a research project using sequential game models to identify what Taiwan can learn from Russia's invasion of Ukraine. Modeling this real-life problem was a challenging task for me. This project required me to use my problem-solving and critical thinking skills as well as history and geography research to model such a complex problem into a model that is simple and comprehensible.
In my research project, I discovered that the dominance of China in all the domains makes the game very unbalanced in favor of China. Thus, most of the strategies that Taiwan can use are defensive. In a war, the defensive strategy tends to be ineffective because the aggressor often knows how the attacked country will react in each scenario. This lack of preparation from the bombed country makes it very difficult for that country to defend itself to win a war. In our sequential game, the tempo is dictated by China. This situation gives China an absolute advantage in that war.
The best strategy for Taiwan is to place all its hope in the overall defense concept. The Ukraine war against Russia showed that this strategy is the best when the aggressor's military power is above the attacked country. This strategy, also known as the "porcupine's strategy" in reference to the animal, allowed Ukraine to gain time and prepare for a long war with Russia. To conclude, the only strategy that can guarantee the sovereignty of Taiwan and a avoid a complete collapse of the country is to use the overall defense strategy as this strategy will give Taiwan a minimum control over the outcome of the invasion." The Attack is the secret of defense; defense is the planning of an attack."― Sun Tzu, The Art of War.
​Data Mining
Sentiment analysis Tool of the New York Times Magazine using Natural Language ToolKit (NLTK)
I coded a sentiment analysis tool specific to the New York Times in the cadre of the case study related to sentiment analysis of the New York Times Magazine using Natural Language Toolkit (NLTK). Using this toolkit was demanding in terms of patience and practice. However, this skill was a game changer for me in conducting some of my following projects. Making the link between qualitative and quantitative dimensions has always fascinated me. I believe that there is a lot under the rock to discover using qualitative data. I also believe that social sciences can benefit a lot by exploring this field.
Economics of Financials Markets
Modern Endowment Management and Embracing ETFs
I was honored to participate in elaborating case studies for undergraduate students related to modern endowments and exchange-traded funds (ETF). My work consisted on generating questions and answers based on the case study. This enriching experience allowed me to understand in detail the subjects I was assigned to and, at the same time, develop my teaching skills. I want to thank my professor Mike Hsu for giving me the opportunity to participate in this project.
Applied Data Science
Research use of logistic models to identify determinants of bankruptcy for Taiwanese companies; selected to present project at University’s annual Graduate Economics Colloquium.
I forged a research project employing data science and data mining tools to identify determinants of bankruptcy for Taiwanese companies. The challenge was related to the financial and legal environment's impact on the definition of bankruptcy as well as the interpretation of the results that followed. The particularity of this project is that there is a plethora of methods when it comes to assessing the potential bankruptcy of a company. However, each method has some limits and can only be applied to certain circumstances.
Furthermore, it is important to study the environment in which a company is before applying any
methods. In our case, in terms of bankruptcy prediction potentiality, the bankruptcy model is more suitable as Taiwanese companies evolve in a debt-oriented environment. Moreover, this model can be used as a proxy or extended with additional ratios to get more accuracy. Finally, this study shows how the environment can change the effectiveness of a model and the need for models which are more dynamic and adaptive to a changing environment.
I successfully exploited and analyzed the data to be consistent with the environment in which these companies evolve; the project was selected to be presented at the Virginia C. Shingleton Memorial Economics Colloquium.
Macroeconomics
​Economics of Developing Nations
Study perspectives, implementation, and challenges of solar energy use as solution for development of growth poles in Burkina Faso
I used my project planning and management skills to conduct a research project related to the perspectives, implementation, and challenges of solar energy use as a solution for developing growth poles in Burkina Faso. The challenge was to find the necessity and the impact of such a project at a micro and macroeconomic scale. The lack of surveys and data related to energy consumption by the population in certain regions required me to develop my creativity and problem-solving skills to investigate further using national articles and to contact some resources to get more information.
The objective of this project is to give as much as details on how the project can be implemented in the province of Bogande. Solar energy is one of the energies of the future. Investing in that energy as our project's intent is to build a path to sustainable development. It is important to note that the beneficial effects of these types of projects are pretty vast and more often observable in the long term, hence the necessity to be sure that the steps in the process are respected to the letter. Then, we will be sure that the projected and expected effects are accurate. I had also the chance to present that project and defend its feasibility.
Applied Data Science
Conference Presentation: Bayesian Analytics
I conducted a research about bayesian analytics that I presented during a conference presentation. The purpose of this project was to show the strengths and the limits of this research methodology when conducting empirical studies. I enjoy this project as this methodology was not used often in class and this presentation was a way for me to master the intricacies of this methodology. In addition, this project allowed me to refresh my skills in probabilities.
Conference Presentation: The Peril and Promise by the Economist
I conducted a research presentation related to economic and monetary policies under the COVID-19. I assess the consequences of monetary policies on the aggregate demand and supply of the U.S as well as the effect of the COVID-19 on the components of the aggregate supply and demand. This presentation was an opportunity for me to make more researches about the aggregate supply which is often less considered compared to the aggregate demand. This presentation was also an opportunity for me to develop my presentation and communication skills.
Microeconomics
Conference Presentation: The secret economics of the Birkin bag, Notion of Veblen and Giffen goods
I conducted a financial statement analysis of Bank Of Africa (BOA) Burkina Faso using vertical and horizontal analysis. The project required me to focus a lot on details as the data needed to be interpreted in conjunction with the environment and the time at which these data were collected. I enjoyed this project as it allowed me to apply my analysis and critical thinking skills to draw pertinent conclusions.
Financial analysis
Conference Presentation: Financial statement analysis of Bank Of Africa(BOA) Burkina Faso using vertical and horizontal analysis
I conducted a financial statement analysis of Bank Of Africa (BOA) Burkina Faso using vertical and horizontal analysis. The project required me to focus a lot on details as the data needed to be interpreted in conjunction with the environment and the time at which these data were collected. I enjoyed this project as it allowed me to apply my analysis and critical thinking skills to draw pertinent conclusions.
Money and Banking
Web app: Developed a web app to automatically predict the Loan Loss Reserve To Total Loans ration using the Federal reserve Beige book and the Goverors' speeches
Applied Econometrics
Conference Presentation: Factors leading to school dropout in Burkina Faso; selected to present project at University’s annual Graduate Economics Colloquium.
I conducted research project examining drivers of school dropout rate in Burkina Faso using Maximum Likelihood Estimation Method and Logistics models like Probit and Logit; I was selected to present the project at Valparaiso University's annual Graduate Economics Colloquium. The challenge was related to some preconception that are often indexed as being the cause of school dropouts. The project was selected to be presented at the Virginia C. Shingleton Memorial Economics Colloquium.
In my research, I found that accessibility to communication technologies allow students in developing countries to easily get access to information which decreases considerably the school dropout rate.
International Economics
Conference Presentation: Immigration restrictionism and Boomerang effect
I conducted a research about the rybczynski theorem and the notion of comparative advantage using the magazine the Economist as a case study. The articles that I used for my research were titled respectively: "the economics of migration, delayed reaction" and "Comparative advantage The boomerang effect". These two articles allowed me to deepen my knowledge in international economics and trade. I also liked this presentation as it was my first conference presentation during my Master program in International Economics and Finance.
Financial diagnosis
Conference Presentation: Risk diagnosis and assessment of the banks Banque de l'Union-Burkina Faso (BDU-BF), Coris Bank International (CBI), and Ecobank International
I performed an assessment and risk diagnosis of the banks Banque de l'Union-Burkina Faso (BDU-BF), Coris Bank International (CBI), and Ecobank International. Real case studies can be very demanding, as putting theory into practice can be challenging. Despite these challenges, I found it interesting to analyze these banks and see the subtle differences between them and how these differences define the performance of each of these banks. This Case study allowed me to develop my modeling and analytical skills as well as my data analysis skills.
The findings of this project is that even though all the banks in Burkina Faso are for the majority commercial banks, their business mix is quite different and this helped to explain why some of them have higher market shares and what type of banking activities are suitable for profitability in the banking environment of Burkina Faso. In addition, I discovered that most of the profits of these banks come from the commissions and fees they make on transactions and products.
