Regression Using Real-World Datasets in R

 

Regression Using Real-World Datasets

Problem 1: House Price Prediction Using Multiple Linear Regression

Problem Statement

A real estate company wants to develop a model to predict the price of a house based on its characteristics. Download a suitable House Price/Housing dataset from Kaggle in CSV format and perform regression analysis using R.

Tasks to be Performed

  1. Download a suitable housing dataset from Kaggle.
  2. Import the CSV dataset into R.
  3. Display the first few records and examine the structure of the dataset.
  4. Identify the dependent variable and suitable independent variables.
  5. Check the dataset for missing values and perform necessary preprocessing.
  6. Select suitable numerical features such as area, bedrooms, bathrooms, parking, or other available variables.
  7. Perform exploratory analysis using summary statistics and suitable plots.
  8. Develop a Multiple Linear Regression model to predict house prices.
  9. Display and interpret the regression coefficients.
  10. Obtain the predicted values and residuals.
  11. Evaluate the model using suitable measures such as:
  • R-squared
  • Adjusted R-squared
  • MAE
  • RMSE
  1. Compare the actual and predicted house prices using a suitable visualization.
  2. Use the developed model to predict the price of a new house with specified characteristics.
  3. Write a brief conclusion based on the results obtained.

Problem 2: Medical Insurance Cost Prediction Using Multiple Linear Regression

Problem Statement

An insurance company wants to estimate the medical insurance charges of customers based on their demographic and personal characteristics. Download a suitable Medical Insurance dataset from Kaggle in CSV format and develop a regression model using R.

Tasks to be Performed

  1. Download a suitable Medical Insurance dataset from Kaggle.
  2. Import the CSV dataset into R.
  3. Examine the dataset using appropriate functions.
  4. Identify the target variable representing insurance charges.
  5. Identify suitable predictor variables such as:
    • Age
    • BMI
    • Number of children
    • Smoking status
    • Other relevant features available in the dataset
  6. Check for missing values and perform necessary preprocessing.
  7. Convert categorical variables into appropriate formats where required.
  8. Perform exploratory data analysis using summary statistics and suitable plots.
  9. Develop a Multiple Linear Regression model to predict insurance charges.
  10. Examine and interpret the regression coefficients.
  11. Generate predicted insurance charges for the observations.
  12. Calculate residuals and analyze the prediction errors.
  13. Evaluate the regression model using:
  • R-squared
  • Adjusted R-squared
  • MAE
  • RMSE
  1. Compare actual and predicted insurance charges using a suitable plot.
  2. Predict the insurance charge for a new customer with given characteristics.
  3. Write a conclusion about the usefulness and performance of the developed model.

Problem 3: Customer Purchase Prediction Using Logistic Regression

Problem Statement

A marketing company wants to predict whether a customer will purchase a product or not based on customer characteristics. Download a suitable Customer Purchase or Social Network Advertisement dataset from Kaggle and perform classification using Logistic Regression in R.

Tasks to be Performed

  1. Download a suitable Customer Purchase dataset from Kaggle in CSV format.
  2. Import the dataset into R.
  3. Display the first few records and examine the structure of the dataset.
  4. Identify the dependent variable representing whether a customer purchased the product.
  5. Identify suitable predictor variables such as:
    • Age
    • Estimated Salary
    • Other relevant customer characteristics
  6. Check for missing values and perform necessary preprocessing.
  7. Convert the target variable into an appropriate binary format if required.
  8. Perform exploratory analysis of the dataset using suitable statistical summaries and plots.
  9. Develop a Binary Logistic Regression model to predict whether a customer will purchase the product.
  10. Obtain the predicted probabilities for each observation.
  11. Convert the predicted probabilities into class labels using an appropriate threshold.
  12. Compare the predicted classes with the actual classes.
  13. Create a confusion matrix.
  14. Calculate suitable classification performance measures such as:
  • Accuracy
  • Precision
  • Recall/Sensitivity
  • Specificity
  1. Use the developed model to predict the purchase decision for a new customer.
  2. Write a brief conclusion based on the classification results.

Comments

Post a Comment

Popular posts from this blog

Statistical Methods Lab ( R Language) PCCBL308 Semester 3 KTU BTech CB and CU 2024 Scheme - Dr Binu V P

  About Me  - Dr Binu V P Statistical Methods Lab ( R Language) PCCBL308 Course Details and Syllabus Introduction to R Download R Download R Studio Tutorial Experiments      1. Familiarization of R environment and R Studio. Installing and using packages.      2.Practice basic R input/output commands and create simple R programs                        using variables/mathematical operations. R Variables Basic Data Types in R Basic Input and Output Functions in R Keywords in R Operators in R Sample Programs Basic Programs to Try - Assignment-1      3.Learn Control Statements Decision-Making Statements in R ( if , if else, switch) Iterative  statements ( for , while , repeat ) Sample Programs using Control Statements Programs to Try using control satements - Assignment-2      4.Write R programs using functions ( Functions, Recursive Functions...
Read more

Programs in R - using control statements - Assignment 2

 Program List 1.Read a mark ( out of 100) and print the corresponding grade. Percentage Range Grade >=90 O >=85 and <90 A+ >= 80 and <85 A > =70 and <80 B+ > =60 and <70 B > =50 and <60 C >=45 and <50 P <45 F 2.Write a program to check the quadrant of a given point(x,y) 3.Write a program that accepts the length of three sides of a triangle as input and determine whether or not the triangle is a right triangle.( university question) 4.Given three points (x1,y1 ) ,(x2,y2) and (x3,y3), check whether they form a triangle. 5.Generate the series 1 2 4 7 11 16....n ( use while...read n) 6.Generate the Fibonacci series 0 1 1 2 3 5 8…..n ( use while..read n) 7.Check whether the given 3 digit number is an Armstrong number. (use while Eg:153,370,371,407) 8.Find the factors of a number ( use for) 9.Find the GCD or HCF of two numbers ( use Euclid algorithm and repeat loop) 10.Check whether the given number is a Krishnamurti number( Krishnamurthy Number...
Read more

Programs to try using Functions in R - Assignment 3

Program List Functions 1.Write a program to compute nCr, using a factorial function 2. Write a function to print the fibonacci numbers between a lower bound and an upper bound 3.Find the binary equivalent of each digit of a number(ie;BCD each digit is represented as 4 bit binary). Use a separate function to find binary. 4.Find the sum of the cosine series 1-x^2/2!+x^4/4!-x^6/6!......n terms.Use a function fact to find the factorial of the number 5.Write a R function is_prime(n), which returns True if the number n is a prime and returns False, if the number n is not prime.Use the is_prime(n) function to generate first N prime numbers 6.Write a pow(x,y) function to compute x^y.Use this function to convert an octal number to decimal. Recursive Functions 1.Write a function to find the greatest common divisor (GCD) of two numbers.(use euclid algorithm) Use this function to find LCM. 2.Write a recursive function to reverse a number. 3.Write a recursive function to find the sum of digits of a...
Read more