NBA Players Salaries Analysis
Team: Jimmy Chan, Cho Fung Chan, Justin Lau, Anthony Dela Paz
Link: See the Github repo
Report: Download
Shiny App(It may take a few minutes for activation):
- Relationship Between Players Statistics
https://statdataviz.shinyapps.io/stat-salaries
- Statistics of NBA Teams
https://statdataviz.shinyapps.io/team-salaries
This project analyzes the data about basketball players from the National Basketball Association (NBA) League in the 2015-2016 season. The central topic is analyzing the salary and performance of NBA players. The motivating questions are how do the skills of a player relate to his salary and the potential overpayment and underpayment of the players. We used efficiency statistics (EFF) to the player in term of his salary to determine the potential overpayment and underpayment of the players. The data reveals that the players are generally overpaid. We also created a shiny app to visualize the relationship between players statistics, e.g., salaries, field goal, rebounds, blocks, etc.
Census Income Prediction
Team: Jimmy Chan, Benny Chen
Link: See the Github repo
Report: Download
The purpose of this project is to build models on the Census Income Data Set. The prediction task is to determine whether a person makes over 50K a year. Using the data from the Census Income Dataset, we wanted to see which tree-based method had the highest predictive accuracy and which variables have the strongest predictive power in regards to whether an individual earns over $50000 in income. We compare three methods - classification trees, bagged trees, and random forest. We begin with preprocessing and exploration of the data. Then, we fit the three models using the training data. With the most accurate model, we fit the test data.