MS in Operations Research, Columbia University in the City of New York
Aug 2015 – Dec 2016
- Machine Learning
- Business Analytics
- Advanced Data Analysis
- Stochastic Modeling
- Probability and Statistics
- Capital Markets and Investments
- Applied Consulting
- Design and Agile Project Management Engineering Lab
- Entrepreneurial Business Creation for Engineers (Teaching Assistant)
BS in Chemical Engineering, Birla Institute of Technology and Science (Pilani)
AUG 2009 – Jun 2013
- Operations Research
- Numerical Analysis
- Linear Algebra
- Advanced C Programming
- Principles of Management
- Principles of Economics
- Financial Management
- Media Advertising
Shirts With Stories
Shirts With Stories is a unique three-sided marketplace designed to facilitate charities to connect with designers for designing merchandise for fundraising. The vision of this product is to help lesser-known charities gain attention and create meaningful, and attractive t-shirts that customers will love while minimizing upfront investment.
The goal for this project was to get hands on understanding of human-centred design process for creating a tech product.
Box Office Analytics
Film industry across the globe is a $90B market. Of that Hollywood contributes about $28B annually. Analytics within the domain is fairly underutilized, making it a lucrative market. The goal of the project was two-fold:
- Making buy-in decisions at pre-production stages using historical data to estimate revenue collections based on early decisions including cast, genre, et al.
- Prediction of revenue collections in pre-release stages. This prediction is based on ratings collected from focus group, demographic data of focus group, data points from movie (including cast, genre, etc)
Future work in this project would include performance of historical movies to optimize marketing mix and audience segmentation for focussed marketing efforts. Final report for the project can be viewed here.
New York City Crime Analysis
In December 2015, New York City Police Department released incident level crime data in NYC for three quarters of 2015. Making use of this data set combined with geographical, temporal, economic and socio-demographic information on census websites, predictive models were created with spatial techniques. The goal was three-fold:
- Modeling historical crime reports for predicting crime rates in future that can be used for deployment of forces. In addition to random forests, poisson families, etc., geographically weighted regression (GWR) techniques were also used for the prediction
- Predicting crime rates in certain new areas based on information available for other areas
- Assisting in relocation based on facilities required (access to schools, stores, pharmacies, etc.)
Results from the analysis are available here.
Stock Price Analysis
Analyzed price movements of tech companies listed in the S&P500. Created a R Shiny app for hypothesis testing of comparison of means and momentum-based prediction in future stock prices.
Worked on analyzing investment in Union Pacific Rail Corporation (SYM: UNP) using fundamental analyses and basic technical tools. The final deliverable was a project report that was presented to the head of Equity Research club.
- Programming: Python, C/C++, Excel VBA
- Statistical Tools: R
- Database Management: SQL
- Data Visualization / Analytical: Tableau, Google Analytics, Mixpanel
- Design / Prototyping: Sketch, InVision, Balsamiq
- Project Management: JIRA, Trello
My codes are available on github.com/parth0708
- Business Strategy
- Growth Hacking
- Data Analysis
- Customer Outreach
- Market Research
- Agile Methodologies