The dataset contains information on 1000+ deliveries by 2 individuals who spent a few months working part-time. My intention was to build a predictive model based on the available variables to estimate the tip percent, but as I progressed my analysis I wondered, must I always build a model ? Or can I always build a predictive model ? Check out my journey on realizing a critical fact that every data analyst/scientist must come to terms with - Do not rush into building a model! Sometimes, you have to live with a baseline model.
Information about the files: Tip-dataset.csv: The dataset is sourced from a fellow Github user with an ID angelddaz. The dataset has more 1290 records. Metadata.png: This image file describes all the fields in the dataset. tipMePlease.rmd: An R-markdown file which contains the analysis and its corresponding code. Figs: This folder contains all the charts, graphs that one could skim through to get a hang of what is done in the project.