Linear Regression
A way to model the relationship of a one or more explanatory variables to a dependent variable
Types Of Linear Regression
- Simple Linear Regression - One explanatory variable
- Multiple Linear Regression - More than one explanatory variable
Simple Linear Regression
`y = beta_0 + beta_1 x`
- y - response variable
- x - explanatory variable
- B0 - intercept
- B1 - slope
Pearson’s Correlation
Correlation is a linear association between two scalar variables
R or Pearson’s R - Correlation coefficient
- always between -1 and 1
- no unit
- corr(x, y) = corr(y, x)
- magnitude indicates strength
- sign indicates direction of association
Residuals
A residual is difference between observed and predicted values
`e_i = y_i - hat y_i`