Module 1.1 Intro to Linear Regression Flashcards
measure of strength of the linear relationship between two variables
correlation coefficient (r)
has no unit of measurement
r (correlation coefficient)
r (correlation coefficient) =
covariance of X and Y/(std dev of X)*(std dev of Y)
what is correlation coefficient bounded by
1 and -1
slope coefficient (b-hat 1) =
covariance of X and Y/variance of X
represents the value of the dependent variable at the point of intersection of the regression line and the axis of the dependent variable
estimated intercept (b-hat 0)
occurs when variance of residuals differ across observations
heteroskedasticity
if observations are NOT independent, residuals from model will exhibit what?
serial correlation
what are the two necessary assumptions of linear regression analysis?
residuals are normally distributed and constant variance of error term