STAT 2080 Lecture Notes - Lecture 9: Dependent And Independent Variables, Scatter Plot

• School: University of Guelph
• Department: Statistics
• Course: STAT 2080
• Professor: Gerarda Darlington

Document Summary

Does not distinguish between outcome and explanatory variables. R not affected by units of x and y. R only measures strength of linear relationship. R close to 0 indicates weak linear relationship. 16 healthy adults overfed for 8 weeks. Increase in energy use based on nea (cal) measured. Regression line is mathematical model for linear relationship. Regression line: summarizes how a change in explanatory variable, x, leads to a change in outcome variable, y. Can use line to predict values for y for given values of x. B1 is the slope: the amount y changes when x increases by 1 unit. Predicting far outside the range of observed x. R2 is the fraction of the variation in y that is explained by the least squares regression of y on x. Least squares regression line of y on x passes through (x bar and y) Regression completely described in terms of xbar, sx, ybar, sy and r.