Welcome!

Who am I?

Hi, how are you?
My name is Ana, I graduated in statistics from the Universidade Federal Fluminense - UFF. I also have a specialization in Data Science from the Pontifícia Universidade Católica do Rio de Janeiro - PUC Rio. I love Harry Potter , Beyoncé, chocolate, my four-legged sister Flor, data analysis, and mathematical modeling! But in this space, I'll be talking about the last two topics. Down below, you'll be able to see a little more about the work I've done and some books I've read that I find interesting to share 😊

Projects

Here, I've separated some of the work I've done.

Beyoncé Spotify

I used the Spotify API to collect some information related to Beyoncé playlists for all the countries where she had concerts in the last tour and Brazil to emphasize how engaged the BR fans are.

Read more

BeesTalk

My friend Ana Santos and I discussed (in Portuguese) briefly on BeesTalk the use of the mean and the interpretation of data as a key factor in problem-solving.

Read more

Credit Score

Credit Score model developed during my internship at BNDES

Read more

ggplot2 in R

A brief demonstration of how we can use ggplot2 in R

Read more

Python API

Creating an API using Python for querying a risk model

Read more

Sampling function in R

Creation of an R function that extracts a subset from a dataset

Read more

Linear regression

Adjustment of a linear regression model to assist in revenue prevision

Read more

Categorical variable biostatistics

Biostatistical analysis using odds ratio, p-value, chi-square test, and logistic regression

Read more

Book recommendations

I've compiled here some books that I've read and have greatly contributed to my learning!

Storytelling with data

The book explains in an easy-to-understand language how we can tell better stories in our presentations and make it clear what we want to convey/recommend through data. One of the things I liked the most was the tips on how to improve charts, both by removing unnecessary elements that only create clutter and by focusing the audience's attention on pre-attentive attributes. After all, I often create a chart that is clear to me but may not be intuitive for my audience.

How to Lie with Statistics

This book was published in the 1950s and brings various reflections on how often data is presented in a convenient way for the presenter, meaning that manipulations are made to disclose a 'desirable' result. These manipulations can be complex or simple, such as altering the y-axis range on a graph to give the illusion of significant growth/decline. Even today, and I believe forever, we will see manipulated statistics in advertisements and news articles that immediately steer public opinion in a certain direction. The book raises some questions we should ask ourselves when we encounter certain analyses, for example: How does the person know this? Does this make sense? This book illustrates everything we shouldn't do in data analysis in a clear and humorous way.

Credit Scoring

Written by the master in statistics Abraham Laredo, who is one of the pioneers in credit risk models for the banking market in Brazil. This book covers all the necessary steps to develop a credit risk model and also demonstrates how to study the accuracy of this model. For those interested in developing in this area, this is the ideal book!

Talk to me!

Want to talk about something you saw here? Send a message, and I'll respond quickly!

Message sent!

Your message has been sent successfully. I promise to respond to you very soon =)

Ana Andrade