ML-Exercise-Advertising-Simple-Linear-Regression

Dataset - https://www.kaggle.com/code/ashydv/sales-prediction-simple-linear-regression/input

Advertising Dataset Analysis

Project Overview

This project involves analyzing Kaggle's Advertising Dataset to build a simple linear regression model that predicts sales based on advertising spend. The dataset includes columns for TV, Radio, Newspaper advertising budgets, and Sales.

Author

• Name: Himel Sarder
• Email: info.himelcse@gmail.com
• GitHub: Himel-Sarder

Objective

The goal of this project is to:

1. Explore the relationship between advertising spend on TV and sales.
2. Build a linear regression model to predict sales based on TV advertising.
3. Evaluate the model's performance using statistical metrics.

Dataset

The dataset contains the following columns:

• TV: Advertising budget for TV (in thousands of dollars).
• Radio: Advertising budget for Radio (in thousands of dollars).
• Newspaper: Advertising budget for Newspaper (in thousands of dollars).
• Sales: Sales generated (in thousands of units).

Sample Data

TV	Radio	Newspaper	Sales
230.1	37.8	69.2	22.1
44.5	39.3	45.1	10.4
17.2	45.9	69.3	12.0
151.5	41.3	58.5	16.5
180.8	10.8	58.4	17.9

Steps

1. Import Libraries

The following Python libraries are used:

• warnings to suppress warnings.
• matplotlib.pyplot and seaborn for data visualization.
• pandas and numpy for data manipulation.
• math for mathematical operations.
• sklearn for machine learning tasks.
• statsmodels for statistical modeling.

2. Load the Dataset

The dataset is loaded using:

import pandas as pd
df = pd.read_csv('advertising.csv')

3. Exploratory Data Analysis (EDA)

• Scatter plot to visualize the relationship between TV advertising and Sales:

plt.figure(figsize=(12, 8))
plt.scatter(df['TV'], df['Sales'])
plt.xlabel('TV')
plt.ylabel('Sales')
plt.show()

4. Data Preparation

• Features (X) and target (y) are defined as:

X = df.iloc[:, 0:1]  # TV column
y = df.iloc[:, -1]   # Sales column

• Split the dataset into training and testing sets:

from sklearn.model_selection import train_test_split
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=100)

5. Model Training

• A linear regression model is trained using:

from sklearn.linear_model import LinearRegression
lr = LinearRegression()
lr.fit(X_train, y_train)

6. Model Evaluation

• The model's performance is evaluated using:
  • Mean Squared Error (MSE)
  • Root Mean Squared Error (RMSE)
  • R-squared

from sklearn.metrics import mean_squared_error
predictions = lr.predict(X_test)
rmse = sqrt(mean_squared_error(y_test, predictions))
print(f"RMSE: {rmse}")

• The regression equation is printed:

print(f"The linear model is: Y = {lr.intercept_:.5f} + {lr.coef_[0]:.5f}X")

7. Statistical Analysis

• The model is refitted using statsmodels for detailed statistical insights:

import statsmodels.api as sm
X_train_sm = sm.add_constant(X_train)
lr_sm = sm.OLS(y_train, X_train_sm).fit()
print(lr_sm.summary())

8. Error Analysis

• Residuals are analyzed to ensure the model's assumptions are met:

res = y_train - lr_sm.predict(X_train_sm)
sns.distplot(res, bins=15)
plt.title('Error Terms')
plt.xlabel('y_train - predictions')
plt.show()

Results

• The regression equation is:

  Y = 6.94868 + 0.05455X
  

• Key performance metrics:
  • RMSE: 2.019
  • R-squared: 0.816
• The model explains approximately 81.6% of the variance in sales based on TV advertising.

Conclusion

• TV advertising has a significant positive relationship with sales.
• The linear regression model is effective in predicting sales based on TV advertising spend.

References

• Kaggle's Advertising Dataset
• Python libraries: sklearn, statsmodels, matplotlib, seaborn, pandas, numpy

---

For more details, visit GitHub.