Predicting Customer Spending

using linear regression analysis
Project Overview
Understanding what drives consumer spending is key to making data-driven business decisions. In this project, we analyzed the relationship between household size, income, and the amount charged using linear regression models, testing for statistical significance and predictive accuracy.
Key Insights & Findings
  • Household Size & Spending: For every additional household member, expected spending increased by $404.13.
  • Income & Spending: For every $1,000 increase in income, expected spending increased by $40.48.
  • Best-Fit Model: A multiple regression model using both household size and income provided the most accurate predictions (Adjusted R² = 0.818), significantly outperforming single-variable models.
My Contributions
  • Data Cleaning & Preparation – Organized and structured raw data for accurate analysis using Excel/XLSTAT.
  • Insights & Business Applications – Developed key takeaways, translating the statistical significance of our findings to use in predicting consumer behavior.
  • Report Writing & Presentation – Translated complex statistical findings into clear, business-focused insights, for ease of decision-making.
  • Regression Modeling – Built and tested multiple linear regression models to determine the impact of household size and income on spending.
Methodology & Tools
  • Simple & Multiple Linear Regression
  • Hypothesis Testing (p-values & confidence intervals)
  • Adjusted R² to evaluate model accuracy
  • Excel & XLSTAT for ANOVA and regression statistical analysis
LINKEDIN