Academic Projects
The following subset of projects were done during my time at Carnegie Mellon University
Visualizing taxi cab Usage in New York City
This project was for 36-315, Statistical Graphics and Visualizations. We designed an interactive dashboard to understand different New York City Taxi cab usage events for the 2016 calendar year. The animated sample displays the visualizations created when looking at hourly data. We can look at pick ups and trip distance for different taxi cab events per hour in a more hands-on approach.
Analyzing London Mortality rate
This project was for 36-618, Experimental Design & Time Series. We implemented several time series models (Time Series regression, Vector Autoregressive, and Neural Networks) to understand the daily mortality rate in London from 2002 to 2007. Click the buttons below to see more of my work on this project!
CALCULATING SPEED USING ANOMALY DETECTION
This project was for 36-650, Statistical Computing. I implemented the Isolation Forest algorithm to determine frames in a video that contained anomalies. With the anomaly information I was able to reconstruct the overall velocity of all objects within a video. Click the button below to see the code written for this project on GitHub!
Determining IDEAL LOCATIONS FOR WIND TURBINES
This project was for 36-667, Special Topics in Statistics: Data over Space and Time. I was tasked with analyzing the ideal location to place a wind turbine that would maximize the yield of electrical power. I was given data of wind speeds for 15 points on a grid (seen left) over the Massachusetts coast. Using techniques like Kriging I located the optimal point that would yield maximum potential energy. Click the button below to see the written report for this project!
examining SOCIO-ECONOMIC EFFECTS ON PER-CAPITA INCOME
This project was for 36-617, Applied Linear Models. I was tasked with exploring the effect of a county’s economic and social variables had on the average income per person. I developed several multivariate linear regressions to model complex relationships and eventually determined the “best” model that predicts per-captia income indicated by the data. Click the button below to see the written report for this project!