bruno's picture

Bruno Gasparotto Ponne

Hello! I'm Bruno.

As a legislative advisor at the Federal Senate of Brazil, I am passionate about data exploration as a tool to inform legislative proposals. I am a firm believer in evidence-based public policy and see it as a primary driver to shape a progressive society. To get relevant insights for my work, I analyze data in R and Python.

During my master's degree, I had the privilege of working at the Hertie School Data Science Lab in Berlin. In my role as a teaching assistant, I provided consultations on R, data visualization, and causal inference, further deepening my understanding and skill set in these areas.

In my free time, apart from exploring data, I like exploring new places and cultures. Reading, gaming, history and jogging are also my hobbies. I am also the proud creator of Coding the Past, a blog designed to educate other social scientists about data science.

In 2023, I published my Master thesis Better Incentives, Better Marks: A Synthetic Control Evaluation of the Educational Policies in Ceará, Brazil. This piece of work captures my views about education policies and their implications. Below you can find more about my projects and research.


  • Educa-se: more data, more education

    A dashboard created in R Shiny to monitor and analyze educational data of municipalities in Sergipe, Brazil. The aim of this platform is to provide quality data and analyses to guide mayors in their local educational policies. It was developed as part of my work at the Federal Senate of Brazil during 2022.
    image related to the project
  • Predicting the price direction of Petrobras stock - PETR4

    In this project, I use Random Forest and AdaBoost models to predict if a stock price will increase or decrease in the next hour based on data of the last hour. It was presented as the final paper in the Specialization Course in Machine Learning at PUC Minas University, in 2022.
    image related to the project
  • The Impact of Institutional Arrangements on Student Achievement: Evidence from Brazil

    This project was actually my Master thesis presented at the Hertie School in 2021 and tutored by Professor Christian Traxler. Two causal inference methodologies - fixed-effects and synthetic control method - were employed to examine the effect of the educational reforms implemented in Ceará, Brazil.
    image related to the project
  • Predicting student loan default in Brazil

    This project was presented in the Machine Learning Course at the Hertie School in 2021, tutored by Professor Slava Jankin and Hannah Bechara. It was developed in partnership with two other students*. Our goal was to employ machine learning algorithms to predict loan default probabilities in the context of a Brazilian student loan program called FIES. We employed 5 methods: logistic regression, decision tree, random forest, linear support vector classification, artificial neural networks - ANN, and an ensemble model. The ensemble model achieved the highest area under the curve (AUC) scores. We also discussed the ethical, social, and economic implications of the results obtained.
    image related to the project
  • Alethea

    Alethea's objective is to create a web application to aggregate data on fake news from fact checking Twitter accounts. The goal of the app is to automate the collection of information in one place helping people to be informed in a fast and reliable way.
    image related to the project
  • Can institutional arrangements improve student performance?

    In 2007, a series of reforms were implemented in the education system in Ceará, Brazil. This research aims to explore data on the education quality index to find out if it suggests these reforms had an impact on student achievement. In order to do so, panel data is analyzed through spatial plots, time series and scatter plots.
    image related to the project