https://github.com/nikdata/portfolio
Data Science Portfolio
https://github.com/nikdata/portfolio
Last synced: 4 months ago
JSON representation
Data Science Portfolio
- Host: GitHub
- URL: https://github.com/nikdata/portfolio
- Owner: nikdata
- Created: 2019-08-20T18:19:05.000Z (almost 7 years ago)
- Default Branch: master
- Last Pushed: 2022-02-20T16:23:49.000Z (over 4 years ago)
- Last Synced: 2025-02-22T19:27:47.196Z (over 1 year ago)
- Size: 6.85 MB
- Stars: 0
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Nikhil's Portfolio
This repo holds project reports and my resume for my data science work.
- [Resume](https://github.com/nikdata/portfolio/tree/master/Resume) holds the latest copy of my resume in PDF format
- [Reports](https://github.com/nikdata/portfolio/tree/master/Reports) holds project reports for my academic work at Northwestern University.
---
Here's a brief summary of the projects:
- Wins per game predictive modeling for over 100 years of baseball data using regression and decision trees using *glm*, *caret*, *broom*.
- Wine case purchase volume predictive modeling using poisson, negative binomial, and hurdle models using *glm*, *caret*, *broom*
- Autocrashes predictive modeling using *ordinary least squares (OLS)*, *logistic regression*
- Discrete choice experiment modeling using *Hierarchical Bayes Multinomial Logit* to select product design
- Model to predict target market for campaign using *random forests* and *naïve bayes* models
- Customer segmentation modeling using *hierarchical agglomerative clustering* and *k-means* followed by market segmentation profiling
- Time series forecasting for the DengAI, disease spread competition, models using *xts*, *forecast*, and *neural networks*
- Model to predict donation amounts for a not-for-profit marketing campaign using a variety of machine learning modeling approaches including *boosting*, *bagging*, *random forest*, and *PCR* using caret