Regression and Correlation

Question 1

how do you use algebra to find the least squares coefficients a and b?

Answer 1

use total variability of x and y and find sample variances of x and y (learn equations)

Question 2

uses of the line of best fit

Answer 2

shows rate of change

makes predictions - be careful not to extrapolate beyond the range of the data

Question 3

excel line of best fit

Answer 3

data analysis, regression on a table of data

make a scatterplot and right click to add trendline and R squared

Question 4

correlation

Answer 4

how well a trend line fits (r)
r less than 0 is negative association
r more than 0 is positive association

• makes no distinction between response and explanatory variables
• both variables must be quantitative
• r values are between -1 and 1
• point near 0 are weak
• affected by outliers

has equations involving S