• a statistical method used to predict the relationship between 2 continuous variables
• Goal is to find the best fitted line for a set of data points

Independent Variable (predictor) : x

Dependent Variable (outcome) : y

\(\hat{y_{i}}=b_{0}+b_{1}x_{i}\)

• Simple Linear Regression calculates this equation with b0 being the y-intercept and b1 being the slope of the line.
• This line is calculated so that it minimizes the error for all the data points

• use the built in function lm(x ~ y, data = dataset)
• save the result into an object
• example : using mtcars built in data set, we can use linear regression to predict the cars mpg based on the horsepower

cars_example = lm(mpg ~ hp, data = mtcars)

cars_example now contains the following components :

• coefficients (intercept, slopes)
• residuals (differences between observed values and predicted values)
• fitted.values (predicted values calculated)
• residual (residual degrees of freedom)
• call (exact formula used to run the model)

Use the coefficients of the lm() component to plot the line.

g = g + geom_abline(
  intercept = cars_example$coefficients[1], 
  slope = cars_example$coefficients[2])