Short bio
Hello! π
I am Stelios, or Stylianos, as my ID insists. I graduated - survived from the Department of Statistics and Insurance Science, completing all 47 courses. Even though I didnβt receive the legendary permit A38, I did get my completion certificate, and I am moving forward!
I have a strong passion for data analysis, machine learning, and more recently, deep learning. Right now, I am looking for my first job in Statistics or Data Science, either in a Junior position or through an internship. No matter when this small miracle happens, I will be here doing what I love. Until recently, I was heavily focused on R, but since I need to find a job, I have started learning Python more systematically. During my studies, I completed many assignments using LaTeX, which is a tool similar to Word but mainly used for work with heavy mathematical and statistical notation. In addition, in my free time, I enjoy creating various visualizations, writing analysis articles, and building statistical web applications like dashboards and ML apps. Many of these can be found on my website.
Happy reading!
Studies
A short description of my studies
BSc in Statistics and Insurance Science
University of Piraeus β’ 2025
Indicative courses:
- Statististics: Estimation Theory and Hypothesis Testing
- Probabilities
- Regression Analysis
- Stochastic Procedures
- Non Parametric Statistics
- Sampling Theory
- Statistical Packages (R / SPSS)
Skills
Programming
I have the most experience working with R, having used it extensively across various academic projects and coursework. I have been using it since 2018, and over time my proficiency and overall confidence with the language have grown significantly. To this day, it remains my primary choice for tasks involving data analysis, data visualization, and machine learning. More recently, I have also begun learning Python, as its strong presence in the data science ecosystem particularly in deep learning makes it impossible to overlook. I progressed to a comfortable working level relatively quickly, although I am still continuing to expand my knowledge. That said, I still consider R to be an exceptionally powerful and elegant language for data analysis and machine learning.
Tools
As primarily an R programmer, the main packages I rely on are the meta-packages {tidyverse}, {tidymodels}, and {tidyverts}. These form the core of my workflow for data manipulation, modeling, and time series analysis. Depending on the complexity and requirements of a project, I may also incorporate additional packages such as:
{modeltime}{rvest}{RSelenium}{highcharter}{Shiny}{gganimate}
About this website
My first attempt at building a personal website was in 2022, using Distill, a well-known R package that made it easy to create simple websites. Unsurprisingly, it was a popular tool among people already familiar with R. Coincidentally, around the same time Quarto was introduced, and I migrated to it a few months later. Since then, my website has been built entirely with Quarto. An important addition worth mentioning is the babelquarto package, which allows the website to support two languages (Greek and English). Because of this bilingual setup, I needed to choose a typeface that supports both Greek and Latin characters so that the visual style remains consistent when switching languages. For this reason, the website uses Gentium Plus throughout.
During its first year online, the website used the free domain provided by GitHub, which was stesiam.github.io. Some older material on the site especially in the chart collection may still reference this domain in visualizations from 2022 or 2023. In 2023, I decided to give the site a more professional identity and purchased my current domain. The website itself continues to be hosted for free through GitHub Pages, meaning the only recurring cost is the annual domain renewal.
Most articles on the website are built using R and its ecosystem of packages. I typically rely on packages from the tidyverse for data import, transformation, cleaning, and visualization. Recently, I have also started converting charts in my articles from ggplot2 to highcharter, a visualization library that provides improved interactivity across both desktop and smaller devices. Highcharter also offers additional options for improved accessibility.
Finally, I try to improve the accessibility of the website as much as possible by following basic principles of accessible and inclusive design. To support this, I have added alternative text (alt text) to many images and charts so that the content can also be understood by users who rely on assistive technologies such as screen readers. For the main charts included in my articles, I also try to place a data table alongside the visualization. This allows readers to access the underlying data directly rather than relying solely on the visual representation. These tables are generated using the gt package, which produces semantic HTML tables tables built using properly structured HTML elements. This approach improves compatibility with screen readers and other assistive technologies. Additionally, for some documents currently available in PDF format, I plan to provide an HTML version as well, making the content more reliably accessible across different devices and reading tools.