Noah Brannon Data Portfolio

Welcome to my Portfolio Website!

Hello! I'm Noah Brannon, and I appreciate you taking the time to explore my portfolio. I graduated from the University of Michigan - Dearborn in December 2022, earning a degree in Economics and a minor in History. Currently, I reside in Trenton, Michigan, which is located just 20 minutes south of Detroit. Ideally, I am looking for opportunities in Metro Detroit, but I am open to relocating within the United States. This portfolio showcases data projects, that I have utilized Python, SQL, MS Excel, data visualization tools like Tableau, and CI/CD tools like Docker and Terraform for use in data analysis, data engineering, and machine learning. The majority of these projects revolve around the realms of economics and finance, as those are the areas in which I feel most confident and adept. Last updated 8/19/2024.

Projects

Housing Prices Forecasting

This machine learning project uses economic data to predict real estate price movements. The dataset includes various economic indicators such as mortgage rates, consumer sentiment, inflation rates, housing supply, and stock market performance. The machine learning model used for prediction is a RandomTreeClassifier, which has shown an accuracy rate of 71%. Backtesting techniques are applied to validate the model and assess its performance. Backtesting involves testing the model on historical data to evaluate how well it would have performed in the past. By analyzing these economic factors and using machine learning techniques, the project aims to provide insights into real estate market trends and assist in making informed decisions for buying or selling properties.
Python Code - GitHub Link
Local Database Creation and Data Ingestion via Docker Project

Established a fully functional PostgreSQL database on a local Docker container and successfully ingested and processed a dataset (CSV via API) with over a million data points using Python scripts (Pandas and SQLAlchemy libraries) within the Docker environment
Project Folder - GitHub Link
Life Expectancy Web App

Developed an interactive world map visualizing life expectancy using Flask (backend framework) to connect to a local instance SQLite database and to retrieve data. Utilized Pandas for data processing and manipulation. Streamlined development through Folium a Python library to create Leaflet maps (JavaScript, HTML, CSS) and JSON data manipulation, to create the web application and tooltips
WebApp - Link Code - GitHub Link

GDP API Web Scraping and Predictive Modeling

In this project, I connected to the Bureau of Economic Analysis API to scrape the GDP data for each state. I collected GDP information from the years 1998, 2002, 2007, 2012, 2017, and 2022. Using this data, I developed a model to predict the GDP for each state in 2027. It is worth noting that the model's projections may seem exaggerated since the selected data did not account for any recessions or economic downturns. As each year used was before or after any of the three recessions that occurred in the United States during the timeframe. Consequently, the model assumed a continuous upward trend in GDP. However, while the results may appear ambitious, they are not entirely unrealistic, and detailed explanations justifying these projections can be found in the README file.
Python Code - GitHub Link
Automotive Data Analysis and Insights Project

Conducted SQL (T-SQL) queries on an automotive dataset from Kaggle, utilizing advanced techniques (joins, CTEs, window functions, aggregations) to identify countries, manufacturers, and top models. Extracted key insights, such as average prices and mileage by category. Presented findings in an interactive Tableau dashboard
SQL Queries - GitHub Link Tableau Public - Link
Economic Indicators Statistical Analysis

The project explores the relationships between the Consumer Price Index (CPI), the Unemployment Rate, and the Federal Funds Rate (FFR) in the United States from 1960 to 2022. It was entirely done within R, including the analysis, explanations, and graphs. Its main objectives include examining the validity of the Phillips Curve theory, analyzing the relationship between CPI and the FFR in the context of the Federal Reserve's efforts to control inflation, and exploring the correlations between CPI, Unemployment Rate, and FFR using statistical analysis techniques. The project presents various visualizations, summary statistics, correlation plots, and linear regression models to illustrate the findings. All data was obtained from the Federal Reserve Bank of St Louis (FRED).
R Code - GitHub Link

NYC Airbnb SQL Data Analysis and Tableau Dashboard Project

This was an in-depth analysis of a dataset comprising Airbnb listings in New York City for the years 2019-2020. Through this analysis, I computed average prices and occupancy days for each neighborhood and created two metrics - price-to-occupancy and price-to-review ratios. Furthermore, I conducted a comprehensive examination of overall metrics for NYC Airbnbs, identifying high-ratio listings and exploring borough averages. I then created three interactive Tableau dashboards for the price, price-to-occupancy and price-to-review metrics
SQL Syntax - GitHub Link Tableau - Tableau Link
Wikipedia Python Web Scraping ETL to SQL Project

Web Scrapped the top 50 worldwide manufacturers by revenue in 2021 and top 100 for 2020 off of Wikipedia using the BeatifulSoup library. Extracted the data into pandas dataframes and then transformed the data into CSV files to be uploaded into SQL Server using SQLAlchemy. Create tables and database schema within Python using SQLAlchemy as well.
Python Code - GitHub Link
Baseball Data Analysis Project

: In this project, I conducted an extensive data analysis of baseball statistics from 1945 to 2015, utilizing Python and Microsoft Excel. Using Python, I performed a comprehensive statistical analysis to explore the relationship between the number of home runs and strikeouts per season, revealing a positive correlation between these variables. To gain further insights, I implemented an ARIMA model to forecast three seasons' worth of wins for each of the 30 MLB teams. Based on these predictions, I simulated playoff-like scenarios using a head-and-tails model that assigned higher weights to top-seeded teams. The season standings and playoffs were visualized and presented using Microsoft Excel
Project Folder - GitHub Link

Financial Data ETL and Automation Project

Built a Python script using the yfinance library to retrieve and transform daily stock data (open, close, high, low) for 18 companies from 2000 to the present (approx. 103,000 data points). Developed an SQL Server database for efficient querying and analysis, uploading data into SQL via Python, and enabling continuous updates with a daily refresh at 4:15 PM EST for real-time insights.
Python Code - GitHub Link
S&P 500 Predictive Modeling

This project showcases a predictive model for the S&P 500 index using machine learning techniques. The model is trained on historical data from 2010 to 2022 and achieves a 57% prediction accuracy for this year's index values. By leveraging the RandomForestClassifier algorithm, the model predicts whether the S&P 500 index will increase or decrease based on closing price, volume, open, high, and low features. Backtesting and precision scores evaluate the model's performance, while the use of rolling averages and trend indicators enhances prediction accuracy
Python Code - GitHub Link
Store Sales SQL and Tableau Project

This SQL project focuses on analyzing a hypothetical Store Sales dataset I got from Kaggle. I created new columns such as revenue, profit margin, and discounted price difference based on existing columns such as profit, quantity, etc. The project calculates various metrics such as average sales, total sales, total revenue, average discount, total quantity, and average order duration per region (East, Central, etc.) and state. It also examines the most common shipping method, popular category purchases, and profitability by category and state. Additionally, it identifies customers with the highest sales, profits, and profit margins. Visualizations of the analysis are available in Tableau.
SQL Queries - GitHub Link Tableau Public Link
SQL Projects

Here are some more SQL projects that I have done.
Repository - GitHub Link
Python Projects

Here are some more Python projects I have done.
Repository - GitHub Link
Excel Projects

Here are some Microsoft Excel projects I have done
Repository - GitHub Link