Superstore Sales Analytics

Probability & Statistics Project

A Flask-based web application for statistical analysis of the Global Superstore dataset.

---

How to Run

1. Install dependencies

pip install -r requirements.txt

2. Make sure superstore.csv is in the same folder as app.py

3. Start the server

python app.py

4. Open in browser

http://localhost:5000

---

Pages

Page	Route	Description
Dashboard	/	KPI cards, category/region charts, yearly trend
Descriptive Stats	/descriptive	Mean, median, CI, histogram, correlation matrix
Probability	/probability	Empirical prob, normality test, binomial, Poisson, QQ plot
Regression	/regression	Multiple linear regression, scatter, residuals, predictor
Raw Data	/rawdata	Searchable, filterable paginated table

---

Dataset

Source: Kaggle — Sample Superstore Dataset
Records: ~9,994 orders
Variables: Order Date, Region, Category, Sub-Category, Product Name, Sales, Quantity, Discount, Profit

---

Statistical Methods Used

• Descriptive stats: median, ,variance, IQR, quartiles
• 95% Confidence Intervals
• Skewness and Kurtosis
• Correlation Matrix (Pearson)
• Shapiro-Wilk Normality Test
• Empirical Probability
• Normal Distribution Approximation (Z-score)
• QQ Plot
• Binomial Distribution
• Poisson Distribution
• Conditional Probability
• Simple and Multiple Linear Regression (OLS)
• R², RMSE
• Residual Analysis