Prediction of financial market's direction: the case of Petrobras (PETR4)

This repository contains the codes developed to predict the price direction of the stock PETR4.

This project is part of the final paper in the Specialization Course in Machine Learning at PUC MINAS University.

The repo is organized as follows:

Module 1: Gathering PETR4 price data and related features (libraries: pytrends and yfinance);
Module 2: Feature Engineering, missing data and outliers treatment;
Module 3: Gathering data from Twitter (tweets related to finance);
Module 4: Dataviz;
Module 5: Data preparation;
Module 6: Training the models, hyperparameters tunning, backtesting and evaluation.

Name		Name	Last commit message	Last commit date
Latest commit History 20 Commits
.ipynb_checkpoints		.ipynb_checkpoints
data		data
Mod1_ObtendoDados.ipynb		Mod1_ObtendoDados.ipynb
Mod2_FeatureEngineering.ipynb		Mod2_FeatureEngineering.ipynb
Mod2b_Tratamento de Outliers no R.R		Mod2b_Tratamento de Outliers no R.R
Mod3_TextData.ipynb		Mod3_TextData.ipynb
Mod4_DataViz.ipynb		Mod4_DataViz.ipynb
Mod5_PreparacaoDados.ipynb		Mod5_PreparacaoDados.ipynb
Mod6a_Training.ipynb		Mod6a_Training.ipynb
Mod6b_TrainingText.ipynb		Mod6b_TrainingText.ipynb
README.md		README.md

Provide feedback