Unformatted text preview: variables, and their correlation r. The least-squares regression line is the line Å· = a + bx with slope and intercept residual/error = observed response âˆ’ predicted response The statistical model for a straight line is y-hat = a + bx. The sign of slope (positive or negative) is the same as the sign of r, the correlation coefficient (positive or negative). "Causation" cannot be established from only one observational study. Since y-intercept is a=y-bar-b*(x-bar), if slope, b, equals zero, then the best predictor for y is y-bar, the mean of the y's. Recall: y-hat = a + bx...
View Full Document
This note was uploaded on 02/27/2012 for the course STAT 121 taught by Professor Patticolling during the Winter '11 term at BYU.
- Winter '11