STAT 201 Chapter Notes - Chapter 3: Dependent And Independent Variables, Linear Regression
10 Jan 2017
Department
Course
Professor
Document Summary
Regression line= predicts value for response variable y as a straight line function of x variable. Y denotes predicted value for y: y= a + bx. Y intercept is predicted value of y when x=0: may not have interpretive value if no observations have x value of 0. Slope, b, is amount y changes when x increases by 1 unit. When slope is negative, predicted y value decreases when x increases. When slope=0, regression line is horizontal; no association. Absolute value of slope describes magnitude of change in y. Residual= prediction errors: prediction errors= the difference between the actual y value and the predicted y value. Positive residual occurs when actual y is large than predicted. Negative residual occurs when actual y is less than predicted. The smaller the absolute value of residual, closer predicted value is to actual (better prediction) For an observation, the vertical difference between the point and regression line is the absolute value of residual.
