MATH 1F92 Lecture 9: Chapter 4.3
15 Oct 2016
School
Department
Course
Professor
Document Summary
Coefficient of determination, measures the proportion of total variation in the response variable that is explained by the least squares regression line. Deviation: how far one value deviates from another value (the difference between one value and another value) In regression: we are concerned with three types of deviations: a) total deviation = y - y . On the graph, the green distance + the red distance. b) explained deviation = ^- y . How far a predicted c) unexplained deviation = y - ^. How far an response deviates from the average response. observed value is from its predicted value. (y - y )2 = (y - ^)2 + (^- y )2. R2 = r2 = known as the coefficient of determination, which is the square of the correlation coefficient. To illustrate each of these types, we will look at a data set of square footage in a house (x) vs. price of a house (y).
