Simple_Linear_Regression

Predicting Delivery Time Using Sorting Time

Step 1 Importing Data

Step 2 Performing EDA On Data

a.) Renaming columns

b.) Checking Datatype

c.) Checking for Null Values

d.) Checking for Duplicate Values

Step 3 Plotting the data to check for outliers

Step 4 Checking the Correlation between variables

Step 5 Checking for Homoscedasticity or Hetroscedasticity

Step 6 Feature Engineering

a.) Trying different transformation of data to estimate normal distribution and to remove any skewness

Step 7 Fitting a Linear Regression Model

a.) Using Ordinary least squares (OLS) regression

b.) Square Root transformation on data

c.) Cube Root transformation on Data

d.) Log transformation on Data

Step 8 Residual Analysis

a.) Test for Normality of Residuals (Q-Q Plot)

b.) Residual Plot to check Homoscedasticity or Hetroscedasticity

Step 9 Model Validation

a.) Comparing different models with respect to their Root Mean Squared Errors

Step 10 Predicting values from Model with Log Transformation on the Data

Building a prediction model for Salary hike

Building a simple linear regression model by performing EDA and do necessary transformations and select the best model using R or Python.

Step 1 Importing Data

Step 2 Performing EDA On Data

a.) Checking Datatype

b.) Checking for Null Values

c.) Checking for Duplicate Values

Step 3 Plotting the data to check for outliers

Step 4 Checking the Correlation between variables

Step 5 Checking for Homoscedasticity or Hetroscedasticity

Step 6 Feature Engineering

a.) Trying different transformation of data to estimate normal distribution and to remove any skewness

Step 7 Fitting a Linear Regression Model

a.) Using Ordinary least squares (OLS) regression

b.) Square Root transformation on data

c.) Cube Root transformation on Data

d.) Log transformation on Data

Step 8 Residual Analysis

a.) Test for Normality of Residuals (Q-Q Plot)

b.) Residual Plot to check Homoscedasticity or Hetroscedasticity

Step 9 Model Validation

a.) Comparing different models with respect to their Root Mean Squared Errors

Step 10 Predicting values from Model with Log Transformation on the Data