└── README.md


/README.md:
--------------------------------------------------------------------------------
 1 | # CAR PRICE PREDICTION MULTIPLE LINEAR REGRESSION
 2 | 
 3 | Implementing Multiple Linear Regression to predict Car Prices
 4 | 
 5 | ## Table of Contents
 6 | 
 7 | - [Introduction](#introduction)
 8 | - [Data Source](#data-source)
 9 | - [Features](#features)
10 | - [Learnings](#learnings)
11 | 
12 | ## Introduction
13 | 
14 | The car company wants to enter a new market and needs an estimation of exactly which variables affect the car prices.
15 | The goal is:
16 |   - Which variables are significant in predicting the price of a car?
17 |   - How well do those variables describe the price of a car?
18 | 
19 | ## Data Source
20 | This data set is taken from Kaggle, the link to which is: 
21 |   - https://www.kaggle.com/datasets/hellbuoy/car-price-prediction
22 | 
23 | ## Features
24 | 
25 | This is a very detailed and explanatory project for beginners who are looking to get their hands dirty with Exploratory Data Analysis, Hypothesis Testing, and Linear Regression.
26 | 
27 | - Exploratory Data Analysis of Categorical and Numerical Variables.
28 | - Using 1 Sample t-test to check if a sample car feature (Average Horsepower in a Sedan) from my dataset is a good representation of the population.
29 | - Using Ordinary Least Squares (OLS) and Normal Equation.
30 | - Using Gradient Descent and Cost Function.
31 | 
32 | ## Learnings
33 | 
34 | The first real project I worked on, taught me several things:
35 | - Getting comfortable with working with a relatively large and noisy dataset.
36 | - Using and showcasing several Python techniques and libraries:
37 |   1. Numpy
38 |   2. Pandas
39 |   3. Scipy
40 |   4. Matplotlib
41 |   5. Seaborn
42 |   6. Scikit Learn
43 | - Using techniques like:
44 |   1. Outlier Detection
45 |   2. Identifying Input Features and Target Variables.
46 |   3. Visualizing different graphs for different variables.
47 |   4. Visualizing relationships amongst different variables.
48 | - Hypothesis Testing:
49 |   1. Deciding which test to conduct.
50 |   2. Defining a significant alpha value.
51 |   3. Comparing the alpha with the p-value and t-calculated with t-critical to draw conclusions about the statistical significance and infer information about the population based on the sample.
52 | - Employing OLS and Normal Equation to build a Multiple Linear Regression Model
53 | - Splitting the dataset into training and testing splits.
54 | - Visualising the predictions and actual values via scatterplot.
55 | - Using Regression Performance Metrics to check the accuracy of our model:
56 |     1. Mean Absolute Error
57 |     2. Mean Squared Error
58 |     3. Root Mean Squared Error
59 |     4. R2 Score
60 |     5. Adjusted R2 Score
61 | - Employing Gradient Descent to build the second Multiple Linear Regression Model.
62 | - Feature Engineering:
63 |   1. Introducing new features from the existing ones to prevent overfitting and save computational resources.
64 |   2. Since introducing new features may increase the chances of introducing Multi-Collinearity in the dataset.
65 |   3. Checking for Multi-Collinearity through Heatmap and Variance Inflation Factor (VIF).
66 |   4. Choosing relevant features for the second model.
67 | - Feature Scaling
68 |   1. Using Z-Score Normalization to scale the input features for a better and more accurate model.
69 | - Using Gradient Descent to build the second model.
70 | - Visualising the predictions and actual values via scatterplot.
71 | - Using Regression Performance Metrics to check the accuracy of our model:
72 |   1. Mean Absolute Error
73 |   2. Mean Squared Error
74 |   3. Root Mean Squared Error
75 |   4. R2 Score
76 |   5. Adjusted R2 Score
77 | 
78 | Below you can find a few snippets of the project
79 | 
80 | ![image](https://github.com/hitesh-hetfield/DS_Projects/assets/151897902/71816b84-6bba-4527-abfe-c08a980c02c3)
81 | ![image](https://github.com/hitesh-hetfield/DS_Projects/assets/151897902/4096e75e-2319-4aea-b93c-4dbdb156411b)
82 | ![image](https://github.com/hitesh-hetfield/DS_Projects/assets/151897902/ce9d06e0-22a3-4f46-8d04-80092b0bc256)
83 | ![image](https://github.com/hitesh-hetfield/DS_Projects/assets/151897902/817487e5-6a78-4e6f-bdd3-90418b913618)
84 | 
85 | 
86 | 
87 | 
88 | 
89 | 


--------------------------------------------------------------------------------