About Me
Hello ! My name is Christian Fernández and I am a recent graduate from Lehman College at CUNY with a bachelors in Economics.
I am particularly passionate about data analytics and machine learning and am largely self-taught in Python, SQL, Tableau, and Excel.
I would describe myself as logical and inquisitive and working with data is something that I believe requires both, making it the perfect career choice for me. I find it to be very satisfying whenever I'm able to derive interesting insights from a messy dataset or when I build a machine learning model and get a great prediction score.
Throughout college, I learned a wide range of economic theories as well as accompanying math such as statistics, calculus, and econometrics. I have also taken some online courses in Python and Machine Learning but what has really helped me develop confidence in my skills is building the projects that are highlighted in this site.
I currently reside in New York and speak English and Spanish. In my free time I enjoy weight lifting, baseball, and documentaries.
Projects
In this section, I ONLY include what I feel are my strongest projects and best show my skills, however I do have many more works, neatly organized and prepared for your viewing, which can be seen on my GitHub account linked HERE or you can click the icon at the bottom of the home page.
Plato's Pizzeria Sales Analysis
Tools Used: SQL (MySQL), Tableau
For this project, I took on the role of a data analyst for Plato's Pizzeria, a Greek-inspired pizza place in New Jersey.
Plato's had been collecting transactional data for the past year, but really hadn't been able to put it to good use. My task was analyzing the data and putting together a dashboard to help them find opportunities to drive more sales and work more efficiently.
Questions answered using SQL:
- How many customers did we have each day?
- Were there any peak hours?
- How many pizzas were typically in an order?
- Do we have any bestsellers?
- How much money did we make this year?
- Can we indentify any seasonality in the sales?
- What was the average total per order ?
- Which pizzas made us the most money ?
GitHub link HERE
SQL Case Studies
Tools Used: SQL (PostgreSQL v13)
For this project, I successfully completed well over 100 queries across 8 comprehensive case studies provided by Danny Ma by way of his 8-week SQL Challenge.
The case studies provide multiple realistic end-to-end project scenarios that represent the sort of work performed in a data analytics role and contain everything from customer segmentation to before-and-after analysis to product funnel analysis!
Domains covered in the case studies include:
- Financial Markets
- Digital Marketing
- People Analytics
- Fast Moving Consumer Goods
- Marketing Analytics
- Retail Analytics
GitHub link HERE
MLB Web Scraper
Tools Used: Python (BeautifulSoup, Requests, Pandas, Numpy), Streamlit
- Created an auto-updating Python script with over 400 lines of code to scrape individual MLB player stats going back to 1983 up until the most current game played today.
- Generated a simple web page for visualizing the scraped data using the Streamlit API, including a downloadable data frame and an EDA section.
- Designed separate sections for hitters and pitchers and added filters for year and team.
GitHub Link HERE
Customer Churn Prediction Model
Tools Used: Python (Sci-Kit Learn, Pandas, Numpy, Matplotlib, Seaborn), Jupyter-Lab, PowerPoint
- Analyzed customer churn data for a telecom company containing 20 different features and 7,032 records.
- Gained further insight using Pandas to determine that 26.6% of customers churned, most of them relatively new, and that Fiber Optic Internet Service was the main reason for this.
- Built and tuned 5 different classification models using the sklearn library to predict likelihood of customer churn as well as which factors were causing it with a final accuracy of 80%
GitHub link HERE
U.S Supply Chain Analysis
Tools Used: Excel
- Used MS Excel to analyze 10,325 supply chain shipments and commodity pricing for HIV/AIDS medication from the U.S. from 2006-2015.
- Leveraged Pivot Tables to put together visualizations for a filterable dashboard containing information on top vendors, receiving countries, total shipments, etc by year.
- Discovered that 89% of all shipments were on time and that generic brand drugs were the most transported with total costs amounting to $47.5 million.
GitHub Link HERE
Contact
Click the button below for a copy of my resume summing all of this into a neat one page document for your records or feel free to message me directly below.
Any and all feedback is greatly appreciated.