M5: Introduction to Regression Flashcards
What does X and Y stand for?
X = Explanatory variable
Y = Response variable
Scatterplot
Plot of Y vs X measured on the same individual
Interpretations (DOTS)
Direction - Positive/Negative
Outliers
Trend - Linear, nonlinear
Strength - Weak/Moderate/Strong
What is the purpose of correlation(r)?
Summarize the direction and strength of the straight line relationship between x and y
Interpretation of r
Strength | 0 - 1
Sign | Direction
Regression line purpose
Line that best fits through the middle of the points
Regression line notation
y = a + bx
Least Squares Regression method
a) Find the line that minimizes the prediction errors
b) Line goes through the middle of the points.
c) Minimize the sum of squared residuals
d) Passes through the point (x-, y-)
Residuals formula
Observed - Predicted
y - y^
Coefficient of Determination (R^2)
Percent of variability in y explained by the linear regression on x
b
r (stdv y / stdv x)
a
a = y_ - bx_