Experiment: Regression Analysis Using a CSV Dataset Downloaded from Kaggle in R

1. Experiment Title

Implementation of Regression Analysis Using a Real-World CSV Dataset Downloaded from Kaggle

2. Aim

To import a CSV dataset downloaded from Kaggle into R, preprocess and explore the dataset, build a regression model, evaluate its performance, and make predictions using the developed model.

3. Objectives

After completing this experiment, students should be able to:

Download a real-world dataset from Kaggle.
Import a CSV file into R.
Explore the structure and contents of the dataset.
Handle missing values.
Select suitable independent and dependent variables.
Build a regression model using lm().
Interpret regression coefficients.
Calculate predicted values and residuals.
Evaluate the model using $R^2$ , Adjusted $R^2$ , MAE, and RMSE.
Visualize actual and predicted values.
Make predictions for new observations.

4. Theory

4.1 Regression Analysis

Regression analysis is a statistical technique used to study the relationship between a dependent variable and one or more independent variables.

Simple Linear Regression

When one independent variable is used:

Y = \beta_0 + \beta_1X + \epsilon

where:

$Y$ = Dependent variable
$X$ = Independent variable
$\beta_0$ = Intercept
$\beta_1$ = Regression coefficient
$\epsilon$ = Error term

Multiple Linear Regression

When two or more independent variables are used:

Y = \beta_0+\beta_1X_1+\beta_2X_2+\cdots+\beta_nX_n+\epsilon

The regression model estimates the relationship between the predictor variables and the target variable.

5. Problem Statement

Download a suitable CSV dataset from Kaggle containing numerical variables.

Import the dataset into R and perform regression analysis to predict a numerical target variable.

Students should perform the following:

Import the CSV file.
Explore the dataset.
Check for missing values.
Select a numerical target variable.
Select suitable predictor variables.
Build a regression model.
Analyze the regression results.
Calculate predictions and residuals.
Evaluate the model.
Visualize the results.
Predict the value for a new observation.

6. Suggested Dataset

A very suitable dataset for this experiment is a House Price Prediction dataset from Kaggle.

Download : https://www.kaggle.com/datasets/yasserh/housing-prices-dataset/data

The dataset should contain variables similar to:

Variable	Description
`area`	Area of the house
`bedrooms`	Number of bedrooms
`bathrooms`	Number of bathrooms
`stories`	Number of floors
`parking`	Number of parking spaces
`price`	Price of the house

Regression Problem

Predict the price of a house based on its area, number of bedrooms, number of bathrooms, and parking facilities.

This makes an excellent real-world Multiple Linear Regression experiment.

7. Software Requirements

R
RStudio
CSV dataset downloaded from Kaggle

8. R Program

The following program assumes that the downloaded file is named:


Housing.csv

and is stored in the working directory.


# -------------------------------------------------------
# Regression Analysis Using a CSV Dataset from Kaggle
# -------------------------------------------------------


# -------------------------------------------------------
# Step 1: Import the CSV File
# -------------------------------------------------------

data <- read.csv("Housing.csv")


# -------------------------------------------------------
# Step 2: Display the First Few Records
# -------------------------------------------------------

cat("First Few Records:\n")

head(data)


# -------------------------------------------------------
# Step 3: Display Dataset Structure
# -------------------------------------------------------

cat("\nStructure of Dataset:\n")

str(data)


# -------------------------------------------------------
# Step 4: Display Dataset Dimensions
# -------------------------------------------------------

cat("\nNumber of Rows and Columns:\n")

dim(data)


# -------------------------------------------------------
# Step 5: Display Column Names
# -------------------------------------------------------

cat("\nColumn Names:\n")

names(data)


# -------------------------------------------------------
# Step 6: Display Summary Statistics
# -------------------------------------------------------

cat("\nSummary Statistics:\n")

summary(data)


# -------------------------------------------------------
# Step 7: Check for Missing Values
# -------------------------------------------------------

cat("\nMissing Values in Each Column:\n")

colSums(is.na(data))


# -------------------------------------------------------
# Step 8: Remove Missing Values
# -------------------------------------------------------

data <- na.omit(data)


# -------------------------------------------------------
# Step 9: Select Required Variables
# -------------------------------------------------------

house_data <- data[, c(
  "price",
  "area",
  "bedrooms",
  "bathrooms",
  "parking"
)]


# -------------------------------------------------------
# Step 10: Build Multiple Linear Regression Model
# -------------------------------------------------------

model <- lm(
  price ~ area + bedrooms + bathrooms + parking,
  data = house_data
)


# -------------------------------------------------------
# Step 11: Display Model Summary
# -------------------------------------------------------

cat("\nRegression Model Summary:\n")

summary(model)


# -------------------------------------------------------
# Step 12: Display Regression Coefficients
# -------------------------------------------------------

cat("\nRegression Coefficients:\n")

coef(model)


# -------------------------------------------------------
# Step 13: Calculate Predicted Values
# -------------------------------------------------------

predicted_price <- predict(model)


# -------------------------------------------------------
# Step 14: Calculate Residuals
# -------------------------------------------------------

residuals_value <- residuals(model)


# -------------------------------------------------------
# Step 15: Add Predictions to Dataset
# -------------------------------------------------------

house_data$Predicted_Price <- predicted_price

house_data$Residual <- residuals_value


cat("\nActual and Predicted Values:\n")

head(house_data)


# -------------------------------------------------------
# Step 16: Calculate R-Squared Values
# -------------------------------------------------------

model_summary <- summary(model)


cat(
  "\nR-Squared =",
  model_summary$r.squared,
  "\n"
)


cat(
  "Adjusted R-Squared =",
  model_summary$adj.r.squared,
  "\n"
)


# -------------------------------------------------------
# Step 17: Calculate MAE
# -------------------------------------------------------

mae <- mean(
  abs(
    house_data$price -
    house_data$Predicted_Price
  )
)


cat("\nMAE =", mae, "\n")


# -------------------------------------------------------
# Step 18: Calculate RMSE
# -------------------------------------------------------

rmse <- sqrt(
  mean(
    (
      house_data$price -
      house_data$Predicted_Price
    )^2
  )
)


cat("RMSE =", rmse, "\n")


# -------------------------------------------------------
# Step 19: Plot Actual vs Predicted Prices
# -------------------------------------------------------

plot(
  house_data$price,
  house_data$Predicted_Price,
  main = "Actual vs Predicted House Prices",
  xlab = "Actual Price",
  ylab = "Predicted Price",
  pch = 19
)


# Add reference line

abline(
  a = 0,
  b = 1,
  col = "blue",
  lwd = 2
)


# -------------------------------------------------------
# Step 20: Plot Area vs Price
# -------------------------------------------------------

plot(
  house_data$area,
  house_data$price,
  main = "House Area vs Price",
  xlab = "Area",
  ylab = "Price",
  pch = 19
)


# -------------------------------------------------------
# Step 21: Predict Price of a New House
# -------------------------------------------------------

new_house <- data.frame(
  area = 5000,
  bedrooms = 3,
  bathrooms = 2,
  parking = 2
)


new_prediction <- predict(
  model,
  newdata = new_house
)


cat(
  "\nPredicted Price of New House =",
  new_prediction,
  "\n"
)
Output
First Few Records:

Structure of Dataset:
'data.frame':	545 obs. of  13 variables:
 $ price           : int  13300000 12250000 12250000 12215000 11410000 10850000 10150000 10150000 9870000 9800000 ...
 $ area            : int  7420 8960 9960 7500 7420 7500 8580 16200 8100 5750 ...
 $ bedrooms        : int  4 4 3 4 4 3 4 5 4 3 ...
 $ bathrooms       : int  2 4 2 2 1 3 3 3 1 2 ...
 $ stories         : int  3 4 2 2 2 1 4 2 2 4 ...
 $ mainroad        : chr  "yes" "yes" "yes" "yes" ...
 $ guestroom       : chr  "no" "no" "no" "no" ...
 $ basement        : chr  "no" "no" "yes" "yes" ...
 $ hotwaterheating : chr  "no" "no" "no" "no" ...
 $ airconditioning : chr  "yes" "yes" "no" "yes" ...
 $ parking         : int  2 3 2 3 2 2 2 0 2 1 ...
 $ prefarea        : chr  "yes" "no" "yes" "yes" ...
 $ furnishingstatus: chr  "furnished" "furnished" "semi-furnished" "furnished" ...

Number of Rows and Columns:

Column Names:

Summary Statistics:

Missing Values in Each Column:

Regression Model Summary:

Regression Coefficients:

Actual and Predicted Values:

R-Squared = 0.5101313 
Adjusted R-Squared = 0.5065026 

MAE = 988779.9 
RMSE = 1307931 

Predicted Price of New House = 6142727 
> source("~/test.R")
First Few Records:

Structure of Dataset:
'data.frame':	545 obs. of  13 variables:
 $ price           : int  13300000 12250000 12250000 12215000 11410000 10850000 10150000 10150000 9870000 9800000 ...
 $ area            : int  7420 8960 9960 7500 7420 7500 8580 16200 8100 5750 ...
 $ bedrooms        : int  4 4 3 4 4 3 4 5 4 3 ...
 $ bathrooms       : int  2 4 2 2 1 3 3 3 1 2 ...
 $ stories         : int  3 4 2 2 2 1 4 2 2 4 ...
 $ mainroad        : chr  "yes" "yes" "yes" "yes" ...
 $ guestroom       : chr  "no" "no" "no" "no" ...
 $ basement        : chr  "no" "no" "yes" "yes" ...
 $ hotwaterheating : chr  "no" "no" "no" "no" ...
 $ airconditioning : chr  "yes" "yes" "no" "yes" ...
 $ parking         : int  2 3 2 3 2 2 2 0 2 1 ...
 $ prefarea        : chr  "yes" "no" "yes" "yes" ...
 $ furnishingstatus: chr  "furnished" "furnished" "semi-furnished" "furnished" ...

Number of Rows and Columns:

Column Names:

Summary Statistics:

Missing Values in Each Column:

Regression Model Summary:

Regression Coefficients:

Actual and Predicted Values:

R-Squared = 0.5101313 
Adjusted R-Squared = 0.5065026 

MAE = 988779.9 
RMSE = 1307931 

Predicted Price of New House = 6142727

9. Explanation of the Important Steps

Step 1: Importing the CSV File


data <- read.csv("Housing.csv")

The read.csv() function imports the CSV file into R as a data frame.

Important: The file name must exactly match the downloaded CSV file name.

Step 2: Exploring the Dataset

Students should use:


head(data)

to view the first few records.


str(data)

to understand:

Variable names
Data types
Dataset structure


summary(data)

to obtain descriptive statistics.

Step 3: Checking Missing Values


colSums(is.na(data))

This checks the number of missing values in every column.

For this basic experiment:


data <- na.omit(data)

removes rows containing missing values.

Step 4: Building the Regression Model


model <- lm(
  price ~ area + bedrooms + bathrooms + parking,
  data = house_data
)

This means:

Predict house price using area, bedrooms, bathrooms, and parking.

The model has the general form:

\widehat{Price} = \beta_0+ \beta_1(Area)+ \beta_2(Bedrooms)+ \beta_3(Bathrooms)+ \beta_4(Parking)

10. Predicted Values

The predicted house prices are obtained using:


predicted_price <- predict(model)

Each house will have:

Actual price
Predicted price

11. Residuals

The residual is:

Residual = Actual\ Value - Predicted\ Value

In R:


residuals(model)

A good regression model should generally have residuals randomly distributed around zero.

12. Model Evaluation Measures

R-Squared

$R^2$ represents the proportion of variation in the target variable explained by the regression model.

For example:

R^2 = 0.75

means approximately:

75% of the variation in house prices is explained by the selected variables.

Adjusted R-Squared

Adjusted $R^2$ is particularly useful in Multiple Regression because it considers the number of predictors.

MAE — Mean Absolute Error

MAE= \frac{1}{n} \sum |Y-\hat{Y}|

It measures the average absolute prediction error.

Lower MAE indicates better predictions.

RMSE — Root Mean Square Error

RMSE= \sqrt{ \frac{1}{n} \sum(Y-\hat{Y})^2 }

RMSE gives greater importance to large errors.

Lower RMSE indicates better predictions.

13. Interpretation of the Actual vs Predicted Plot

The graph contains:

X-axis → Actual house prices
Y-axis → Predicted house prices

The line:

Y=X

represents perfect prediction.

Therefore:

The closer the points are to the reference line, the better the predictions.

14. Result

Regression analysis was successfully performed using a real-world CSV dataset downloaded from Kaggle. The dataset was imported into R, explored, and preprocessed. A Multiple Linear Regression model was developed to predict house prices based on selected features. The model was evaluated using R-squared, Adjusted R-squared, MAE, and RMSE, and predictions were made for new observations.

Search This Blog

Statistical Methods Lab ( R Language) PCCBL308 Semester 3 KTU BTech CB CU 2024 Scheme

Regression Analysis Using a CSV Dataset Downloaded from Kaggle in R