Suicides in Brasil: 2010-2019 (PT-BR)

This is a full analyses of the number of suicides in Brazil ranging from 2010 to 2019. Data were collected and cleaned from DATASUS using the PySUS library from Python. Visualizations and descriptive statistics were obtained using R (specially the ggplot2, dplyr, tidyr, and psych packages). A machine learning procedure was at last implemented aiming to predict the number of nationwide suicides per month using few economic variables. On the machine learning procedure, the final neural network model achieved Mean Absolute Error = 0.3135.

Key questions asked were:

1. How many suicides occured by year? And by month? Was there any noticeable uptrend or downtrend throughout the years?
2. Who died of suicide? Age, sex, race, occupation and other demographics were explored.
3. Where did suicide occur? Where were suicides more common and which Brazilian states had higher reports of suicide?

Also, as a final product, datasets were published in a public platform. All datasets can be accessed here.