Author: Dominik Truchly
Date: December 2024
Language: R (R Markdown)
This project performs a series of statistical analyses on a Child Car Seats dataset, covering hypothesis testing, correlation, regression, and classification techniques. It was completed as part of a statistics course assignment.
data/ChildCarSeats.csv — contains retail sales data for child car seats across different stores, including variables such as Sales, Price, CompPrice, Advert, City, and Expensive.
| Task | Method |
|---|---|
| Task 1 | One-way ANOVA — effect of advertisement type on sales; assumptions checked via Levene's test, Shapiro-Wilk test, and Tukey's HSD post-hoc test |
| Task 2 | Pearson correlation matrix and variance-covariance matrix for Sales, Price, and CompPrice |
| Task 3 | Contingency table and odds ratio analysis — association between City and Expensive |
| Task 4 | Multiple linear regression — Sales predicted from Price and City; model diagnostics and prediction intervals |
| Task 5 | Logistic regression — probability of a car seat being Expensive based on Sales |
| Task 6 | Non-parametric comparison of Price and CompPrice medians using the Wilcoxon signed-rank test |
analysis.Rmd— R Markdown source code with all analyses and explanationsreport.html— Rendered HTML report with outputs, plots, and interpretations
dplyr, tidyr, ggplot2, Hmisc, ppcor, GGally, ggcorrplot, gridExtra, agricolae, car, nortest